- Email: [email protected]
The morphological phenotype of β-amyloid plaques and associated neuritic changes in Alzheimer’s disease
PII: S 0 3 0 6 - 4 5 2 2 ( 0 1 ) 0 0 1 6 9 - 5
Neuroscience Vol. 105, No. 1, pp. 99^107, 2001 ß 2001 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306-4522 / 01 $20.00+0.00
www.elsevier.com/locate/neuroscience
THE MORPHOLOGICAL PHENOTYPE OF L-AMYLOID PLAQUES AND ASSOCIATED NEURITIC CHANGES IN ALZHEIMER'S DISEASE T. C. DICKSON and J. C. VICKERS* Discipline of Pathology, Clinical School, University of Tasmania, 43 Collins St, Hobart, Tasmania 7000, Australia
AbstractöWe have utilised laser confocal microscopy to categorise L-amyloid plaque types that are associated with preclinical and end-stage Alzheimer's disease and to de¢ne the neurochemistry of dystrophic neurites associated with various forms of plaques. Plaques with a spherical pro¢le were de¢ned as either di¡use, ¢brillar or dense-cored using Thio£avin S staining or immunolabelling for L-amyloid. Confocal analysis demonstrated that ¢brillar plaques had a central mass of L-amyloid with compact spoke-like extensions leading to a con£uent outer rim. Dense-cored plaques had a compacted central mass surrounded by an outer sphere of L-amyloid. Di¡use plaques lacked a morphologically identi¢able substructure, resembling a ball of homogeneous labelling. The relative proportion of di¡use, ¢brillar and dense-cored plaques was 53, 22 and 25% in preclinical and 31, 49 and 20% in end-stage Alzheimer's disease cases, respectively. Plaque-associated dystrophic neurites in preclinical cases were immunolabelled for neuro¢lament proteins whereas, in end-stage cases, these abnormal neurites were variably labelled for tau and/or neuro¢laments. Double labelling demonstrated that the proportion of di¡use, ¢brillar and dense-cored plaques that were neuritic was 12, 47 and 82% and 24, 82 and 76% in preclinical and end-stage cases, respectively. Most dystrophic neurites in Alzheimer's disease cases were labelled for either neuro¢laments or tau, however, confocal analysis determined that 30% of neuro¢lament-labelled bulb-like or elongated neurites had a core of tau immunoreactivity. These results indicate that all morphologically de¢ned L-amyloid plaque variants were present in both early and late stages of Alzheimer's disease. However, progression to clinical dementia was associated with both a shift to a higher proportion of ¢brillar plaques that induced local neuritic alterations and a transformation of cytoskeletal proteins within associated abnormal neuronal processes. There data indicate key pathological changes that may be subject to therapeutic intervention to slow the progression of Alzheimer's disease. ß 2001 IBRO. Published by Elsevier Science Ltd. All rights reserved. Key words: Alzheimer's disease, L-Amyloid, plaque, dystrophic neurites, neuro¢laments, tau.
may be a key mechanism in the development of AD, the presence of L-amyloid-containing plaques may represent presymptomatic or unrecognised early symptomatic AD (Morris et al., 1991, 1996; Price and Morris, 1999). In contrast, the relative density of plaques may decrease with advanced stages of the disease, presumably due to clearance by associated glial cell types (Thal et al., 1998). The heterogeneity of plaque morphology has also been noted by a number of early investigators in the ¢eld of AD research (Wisniewski and Terry, 1973; Ulrich, 1985), but the pathological signi¢cance of these plaque isoforms to the disease mechanism is yet to be de¢nitively determined (Armstrong, 1998; Ikeda et al., 1990). This variable plaque morphology has led to the development of a number of di¡erent typing schemes that have been consistently modi¢ed and updated with the advent of more sensitive staining and immunolabelling methods. The form of microscopy used in individual investigations has also been recognised as an important contributor to the ability to accurately describe plaque morphology. The localisation of clusters of abnormal neuronal processes, referred to as dystrophic neurites (DNs), with a subset of the L-amyloid plaques is a common phenomenon in AD (Hof and Morrison, 1994). These neuritic plaques have been considered a pathological correlate of AD dementia (Mirra et al., 1991). Standard immunohis-
The deposition of L-amyloid in the brain is a generally accepted, but not speci¢c, pathological hallmark of Alzheimer's disease (AD), however, there is considerable diversity of opinion regarding the signi¢cance of plaque formation to the aetiology of the disease (Arnold et al., 1991; Arriagada et al., 1992a; Selkoe, 1994; Terry, 1996). For example, there have been inconsistent reports on the correlation between L-amyloid plaque density and various measures of clinical de¢cit, and there is an alternative view that more soluble L-amyloid may be of greater pathological importance than the insoluble L-amyloid that comprises plaques (Vickers et al., 2000). Furthermore, in elderly persons without overt dementia, the presence of plaques in the neocortex is often accepted as part of normal aging (Arriagada et al., 1992b), therefore making the determination of the pathological significance of such structures di¤cult. Other investigators suggest that because cerebral deposition of L-amyloid
*Corresponding author. Tel. : +61-3-62264830; fax: +61-362264833. E-mail address: [email protected] (J. C. Vickers). Abbreviations : AD, Alzheimer's disease ; DN, dystrophic neurite; NF, neuro¢lament ; PBS, phosphate-bu¡ered saline; SFG, superior frontal gyrus. 99
NSC 5014 16-7-01
Cyaan Magenta Geel Zwart
100
T. C. Dickson and J. C. Vickers
tochemical methods and £uorescence microscopy have been previously utilised to e¡ectively analyse the exact complement of epitopes expressed within the clusters of abnormal neurites associated with L-amyloid plaques (Dickson et al., 1988, 1999; Masliah et al., 1993b; Vickers et al., 1994, 1996; Yasuhara et al., 1994; Wang and Munoz, 1995; Su et al., 1996; Saunders et al., 1998; Thal et al., 1998;). Confocal microscopy allows for the detailed study of this association between L-amyloid plaques and DNs. Masliah et al. (1993a), through the use of confocal microscopy, concluded, in their investigation of the alterations that occur in synapses and axons in AD, that many of the tau abnormal neurites commonly localised with L-amyloid plaques were continuous with synaptophysin-positive distended terminals. Others have used confocal microscopy to investigate various aspects of senile plaque morphology in end-stage AD cases (Schmidt et al., 1995; Cruz et al., 1997) and also the relationship between senile plaques and astrocytes (Kato et al., 1998). In the current investigation, confocal microscopy was utilised to accurately quantitate the proportion of various plaque morphological types in both preclinical and end-stage AD cases and to investigate their association with neurochemically de¢ned clusters of DNs. These data provide new information on the link between two of the hallmarks of AD, the L-amyloid plaque and DNs, and further elucidate the sequence of pathology development that ultimately leads to nerve cell degeneration and dementia.
EXPERIMENTAL PROCEDURES
Tissue source and processing Blocks of human brain tissue were obtained from the Sun Health Research Institute (Arizona, USA) as previously described (Saunders et al., 1998). Informed consent for the collection of material was obtained prior to death and tissue use was approved by the Institutional Ethics Committee and is consistent with the Declaration of Helsinki. These blocks of cerebral cortex (superior frontal gyrus (SFG), Brodmann area 9) were immersion-¢xed in 4% paraformaldehyde, cryoprotected and sectioned on a freezing microtome at 40^50 Wm. These samples included ¢ve cases of AD conforming to the CERAD (Consortium to Establish a Registry for Alzheimer's Disease) criteria (Mirra et al., 1991) (74, 74, 83, 88 and 92 years of age), and ¢ve preclinical AD cases (78, 81, 84, 90 and 91 years of age). The post-mortem interval for all cases ranged from 2 to 3 h. No di¡erences were observed in the immunolabelling pro¢le for the various antibodies across varying postmortem to ¢xation intervals or di¡erent ¢xation protocols. The preclinical cases were de¢ned by the lack of overt dementia (although de¢nitive neuropsychological data were not available)
and the presence of widespread L-amyloid immunoreactive plaques in the neocortex, with the distinction that these plaques could not be de¢ned as `neuritic' based on Thio£avin S staining or immunolabelling for PHF-tau or ubiquitin (Vickers et al., 1996). Thus, these preclinical cases do not conform to CERAD criteria for diagnosis of clinical AD (Mirra et al., 1991). These cases do correspond to Braak staging of `3' based on the presence of neuro¢brillary tangles in the entorhinal cortex and hippocampus as well as L-amyloid plaques in the neocortex (Braak and Braak, 1991). Notably, these preclinical cases lacked neocortical neuro¢brillary tangles. Staining To investigate the morphology of L-amyloid plaques, 100-Wm sections of SFG from the preclinical and end-stage AD cases were stained with Thio£avin S [0.0125% in 40% ethanol and 60% 0.01 M phosphate-bu¡ered saline (PBS)] (Sigma, St. Louis, MO, USA) for 3 min at room temperature, protected from the light. After staining, sections were di¡erentiated in two 10-min rinses of 50% ethanol and 50% 0.01 M PBS before ¢nal washes in 0.01 M PBS. Immunohistochemistry The relationship between neuro¢lament (NF)-immunoreactive DNs and L-amyloid was investigated in all cases. Sections in which L-amyloid was to be visualised were pre-treated in 90% formic acid prior to incubation in either primary antibody. A rabbit antibody which recognises all carboxy-terminal variants of L-amyloid (AL1^40 and AL1^42(43)), pan L-amyloid, was combined alternatively with mouse monoclonal antibodies to phosphorylated NFs (SMI312) in both preclinical and endstage AD cases. The co-localisation pattern of cytoskeletal proteins (NF, tau) within individual abnormal neurites was also investigated using double labelling immunohistochemistry by combining the rabbit antibody to 4Rtau (four microtubule binding domains, raised against a non-phosphorylated recombinant protein) with the SMI312 mouse monoclonal NF antibody (see Table 1 for primary antibody details). SMI312 is a `cocktail' of monoclonal antibodies to phosphorylated NFs (middle and high molecular weight subunits) and does not cross-react with tau in AD cases (Dickson et al., 1999). For all £uorescence labelling, mouse monoclonal antibodies were visualised with a horse antimouse secondary antibody conjugated to £uorescein isothiocyanate (FITC; Vector Laboratories, Burlingame, CA, USA; 1:200) whereas the rabbit polyclonal was visualised with a goat anti-rabbit antibody conjugated to biotin (Vector, 1:200) followed by Texas Red avidin D (Vector, 1:200). In all cases, the density (/mm2 ) of L-amyloid labelled plaques was quantitated. A rectangular counting frame and 10U objective were utilised to count spherical L-amyloid-immunoreactive deposits (greater than 20 Wm diameter) in the neocortex from non-overlapping ¢elds, representing ¢ve pia to white matter traverses, examined with a Leitz epi£uorescence microscope. Confocal microscopy Laser confocal scanning microscopy, using an Optiscan F900e krypton/argon system attached to an Olympus BX50 epi£uorescence microscope, was utilised to investigate the morphology of various plaque types and the localisation patterns of DNs
Table 1. Antibody details Code
Type
Reactivity
Dilution
Source
SMI32 SMI312 L-Amyloid (pan) Tau
m mc r r
dephosphorylated NF-M and NF-H phosphorylated NF-M and NF-H All L-amyloid peptides 4R Tau
1:1000 1:1000 1:500 1:2000
Sternberger Monoclonals Inc., Lutherville, USA Sternberger Monoclonals Inc., Lutherville, USA QCB, Hopkinton, USA Dako, Glostrup, Denmark
m: mouse monoclonal; mc: mouse monoclonal cocktail; r: rabbit polyclonal.
NSC 5014 16-7-01
Neuritic plaquesin Alzheimer's disease
within the plaque substructure. Thio£avin S-stained and Texas Red-labelled preparations were recorded simultaneously in separate channels. Representative plaques were optically sectioned at intervals of 1^2 Wm for distances of approximately 20^30 Wm through a plaque. This information was then utilised to create three-dimensional images of the varying plaque types present throughout the cases. Plaques were categorised as either `spherical-di¡use', `dense-cored' or `¢brillar'. Similar sectioning methods were utilised for material double labelled with the antibodies to L-amyloid, however, distances of only 5^10 Wm were sectioned due to antibody penetration limitations. Detailed analysis of the plaque substructure was performed using the Optimate 5.2 program. The relative proportion of each plaque type was determined by examining in detail 100 plaques, throughout all cortical layers, in each case (i.e. 1000 plaques in total). Further to this, the proportion of each of these plaque types that were neuritic, i.e. associated with either NF-labelled (preclinical AD cases) (Dickson et al., 1999) or tau-labelled (end-stage AD cases) abnormal neurites, was also determined in each case. Di¡use amyloid was noted to be highly variable in its size and deposition patterns. Therefore, for the purpose of these investigations, only spherical di¡use deposits with a minimum diameter of approximately 40 Wm were considered. Confocal microscopy was further utilised to perform an analysis of the plaque-associated DNs in the AD cases. The 40-Wm sections double labelled with the rabbit antibody to tau and the mouse antibody to phosphorylated (SMI312) NFs were used for these analyses. Clusters of DNs were optically sectioned at 0.5^ 1 Wm intervals. `Optimate' was again utilised for size measurements. The proportion of NF-immunoreactive DNs that had a tau-immunoreactive core was quanti¢ed by sectioning through 100 non-continuous DNs labelled by SMI312 in each AD case. These counts were made throughout the layers, and included all morphological types of DNs.
RESULTS
Preclinical Alzheimer's disease Staining with Thio£avin S and labelling with antibodies to L-amyloid revealed a number of morphologically distinct plaques. Serial `confocal' sectioning throughout these plaques uncovered key features speci¢c to each plaque type. A subset of plaques had a dense core of L-amyloid, typically surrounded by a ring of indistinct wispy labelling (Fig. 1A). Other plaques lacked a dense central core and were comprised of large numbers of distinct, spoke-like ¢bril bundles that emanate from a central mass. These `¢brillar' plaques had distinct pores
101
and irregularities within their structure (Fig. 1B). Of note, the L-amyloid labelling in the outer regions of both of these plaque types was usually con£uent. Di¡use spherical deposits were more irregular in shape and without distinct edges (Fig. 1C). Layer I of the neocortex was notable for its lack of any form of L-amyloid labelling in preclinical AD cases. Variation between cases was low with regard to patterns of L-amyloid labelling. Quantitation of the proportion of these three plaque types in ¢ve cases demonstrated that approximately half of the spherical L-amyloid deposits were di¡use (average of 53.4% þ 2.2, S.E.M.), with dense-cored and ¢brillar plaques accounting for the remaining 24.6% ( þ 2.2, S.E.M.) and 22.0% ( þ 1.7, S.E.M.) of plaques, respectively (Table 2). The mean proportion of the plaques in these cases that were neuritic, as de¢ned by the co-presence of NFlabelled DNs, was 36.8% ( þ 7.2, S.E.M.). Of the di¡use plaques, 12% ( þ 2.1, S.E.M.) were associated with NFimmunopositive DNs. However, 47.0% ( þ 3.6, S.E.M.) of ¢brillar plaques and 81.8% ( þ 3.4, S.E.M.) of densecored plaques were neuritic (Table 3). The NF-immunoreactive DNs exhibited a variety of morphological types including bulb- and ring-like structures. These structures were labelled by SMI312 but not labelled by the antibody to tau. Optical sectioning through neuritic plaques indicated that the larger bulb-like neurites were generally located within larger pores in the amyloid plaque substructure (Fig. 2A). In the dense-cored plaques, these pores mostly occupied the area directly adjacent to the L-amyloid core, whereas, in di¡use and ¢brillar plaques, they were non-uniformly dispersed throughout the plaque. The larger bulb-like DNs were often continuous with ¢ne processes of normal diameter. Many of these processes appeared to pass through pores in the plaque, immediately terminating with a large swelling (Fig. 2A). The ring-like structures also appeared to be co-located to spaces within the L-amyloid plaque, but these internal subregions devoid of amyloid were smaller in dimension to those occupied by the larger bulbs. Alzheimer's disease The same morphological variants of spherical L-amy-
Table 2. L-Amyloid plaque variants
Preclinical case 1 2 3 4 5 Mean ( þ S.E.M.) End-stage case 1 2 3 4 5 Mean ( þ S.E.M.)
Plaque density (/mm2 )
% Fibrillar
% Dense-cored
% Di¡use
21.9 31.2 10.3 10.3 15.4 17.8 þ 4.0
27 17 22 24 20 22.0 þ 1.7
26 23 23 25 26 24.6 þ 0.7
47 60 55 51 54 53.4 þ 2.2
23.0 42.9 40.8 47.5 48.6 40.6 þ 4.6
58 56 43 50 40 49.4 þ 3.5
22 8 22 19 28 19.8 þ 3.3
20 36 35 31 32 30.8 þ 2.9
NSC 5014 16-7-01
Cyaan Magenta Geel Zwart
102
T. C. Dickson and J. C. Vickers
Fig. 1. (A, i^iii) Series of three optical sections taken through a dense-cored plaque stained with Thio£avin S in a preclinical AD case (83 years). The images were captured at intervals of approximately 14.5 Wm. The total depth of the sectioned plaque was 28.56 Wm. Images show a distinct central core (arrow) of L-amyloid surrounded by a void or clearing and then an outer spherical rim of L-amyloid (arrowhead). (B, i^iii) A similar set of images sectioned through a ¢brillar plaque stained with Thio£avin S in a preclinical AD case (81 years). The images were captured at intervals of approximately 17.5 Wm. The total depth of the sectioned plaque was 34.68 Wm. The ¢brillar plaques showed dense L-amyloid accumulations throughout the plaque structure. In the central regions these accumulations often appeared as spoke-like accumulations (arrow), emanating from a denser central accumulation. Distinct pores within the plaque were obvious in deeper sections. (C, i^iii) A set of sequential confocal images sectioned through a di¡use plaque labelled with an antibody to L-amyloid in an end-stage AD case (74 years). The images were captured at intervals of approximately 2.5 Wm. The total depth of the plaque that was sectioned was 5 Wm. These deposits were generally larger in diameter, however, considerably smaller in depth. Only small changes in the plaque structure were noted in the di¡erent optical sections. Scale bar = 20 Wm (A and B), 50 Wm (C).
loid deposits were present in the end-stage AD cases as those noted in the preclinical material, i.e. di¡use, ¢brillar and dense-cored. L-Amyloid labelling was, however, more intense and spread throughout the cortical layers. In addition, there were approximately twice as many L-amyloid plaques in the neocortex of end-stage cases
relative to preclinical cases (Table 3), and this di¡erence was statistically signi¢cant (P 6 0.01). In end-stage AD, large numbers of di¡use L-amyloid-immunopositive deposits were present throughout layer I with many dense-cored plaques present within layer V. Interestingly, whereas other studies have noted a morphological form
NSC 5014 16-7-01
Neuritic plaquesin Alzheimer's disease
103
Fig. 2 (Caption overleaf).
of L-amyloid deposit consisting solely of a dense central deposit (so-called `compact' or `burnt-out' plaques, reviewed by Armstrong (1998)), confocal analysis in this study determined that these were always associated
NSC 5014 16-7-01
with some degree of a corona of surrounding L-amyloid immunoreactivity, thus conforming to the dense-cored classi¢cation. The prevalence of the three spherical plaque types was notably di¡erent in the AD cases as
Cyaan Magenta Geel Zwart
104
T. C. Dickson and J. C. Vickers Table 3. Proportion (%) of L-amyloid plaque variants associated with dystrophic neurites
Preclinical case 1 2 3 4 5 Mean ( þ S.E.M.) End-stage case 1 2 3 4 5 Mean ( þ S.E.M.)
Fibrillar
Dense-cored
Di¡use
57 38 47 40 53 47.0 þ 3.6
92 73 86 82 76 81.8 þ 3.4
17 6 8 14 15 12.0 þ 2.1
83 86 77 82 83 82.2 þ 1.5
81 75 73 74 79 76.4 þ 1.5
20 28 22 32 19 24.2 þ 2.5
compared to the preclinical AD cases. In ¢ve AD cases, proportional analysis demonstrated that 49.4% ( þ 3.5, S.E.M.) of spherical L-amyloid deposits were ¢brillar, 19.8% ( þ 3.3, S.E.M.) dense-cored and 30.8% ( þ 2.9, S.E.M.) di¡use (Table 2). Analysis of variance demonstrated a signi¢cant interaction e¡ect (P 6 0.01) when comparing the proportion of plaque types in preclinical vs end-stage AD cases, which can be attributed to the shift to a higher relative proportion of di¡use to ¢brillar plaque types from preclinical to end-stage AD cases. In this respect, there was a signi¢cant di¡erence (P 6 0.01) in the mean proportion of di¡use and ¢brillar plaques between the case categories, but interestingly, no signi¢cant di¡erence in the proportion of dense-cored plaques between preclinical and end-stage AD cases. Labelling of Thio£avin S-stained sections with an antibody to tau demonstrated that 76.4% ( þ 1.5, S.E.M.) of dense-cored plaques, 82.2% ( þ 1.5, S.E.M.) of ¢brillar plaques and 24.2% ( þ 2.5, S.E.M.) of di¡use deposits were neuritic (Table 3). The mean proportion of all plaques that were neuritic was 64.0% ( þ 3.3, S.E.M.). The relatively higher proportion of both di¡use and ¢brillar plaques associated with abnormal neurites in end-stage relative to preclinical AD cases was con¢rmed by statistical analysis (P 6 0.01). However, there was no signi¢cant di¡erence in the proportion of neuritic dense-cored plaques in either case type, where the relative proportion of this plaque types associated with DNs was high. The predominant morphological DN types included bulb-like structures of varying sizes and also elongated, fusiform structures (Fig. 2). Both forms exhibited immunoreactivity to anti-tau and anti-NF antibodies. The pat-
tern of distribution of the bulb-like DNs closely resembled that encountered for similar DNs in the preclinical cases. Generally, the bulb-like structures were located within the pores of the plaque (Fig. 2B). The more tortuous and elongated DNs had a somewhat different distribution within the L-amyloid plaques. Rather than occupying pores, these abnormal neurites emerged from small pores and did not appear to be associated with structural features of the surrounding L-amyloid. Consequently, they were located in all areas of plaques including within the central region (Fig. 2B). Analysis of the NF-immunopositive DNs by confocal microscopy con¢rmed the presence of a tau-immunoreactive core within a subset of these pathological structures. Initially this pattern of labelling was thought to be restricted to the bulb-like DNs (Fig. 2C) (Dickson et al., 1999), however, confocal microscopy also revealed tauimmunoreactive cores within the longer, more cylindrical DNs (Fig. 2D). In these structures only the very outer rim showed anti-NF labelling. Confocal investigation of individual NF-immunopositive DNs demonstrated that an average of 30.0% ( þ 4.3, S.E.M.) had a tau core. Many of these cores were only visible after optically sectioning through the DN clusters (Fig. 2E).
DISCUSSION
The current investigation con¢rms previous reports that L-amyloid plaques are not uniform structures (Ikeda et al., 1989; Wisniewski et al., 1989; Dickson, 1997). Both the preclinical and end-stage AD cases
Fig. 2. Double labelling with antibodies to L-amyloid (red) and phosphorylated NFs (green) showing the relationship between plaque morphology and the location of the abnormal neurites in a preclinical AD case (91 years) (A). Arrowheads indicate neuro¢lamentous ring-like structures and arrows, bulb-like structures. (B) Double labelling with antibodies to L-amyloid (red) and NF (green) showing the relationship between plaque morphology and the location of the abnormal neurites in an endstage AD case (92 years) Arrowheads indicate long distended DNs exhibiting classical tau morphology, whereas arrows indicate neuro¢lamentous bulb-like accumulations of similar morphology to those noted in preclinical AD cases. (C) Double labelling of bulb-like NF-immunopositive DNs (green) with a tau core (red) in an AD case (74 years). (D) A similar labelling pattern in an elongated fusiform NF-immunopositive DN (green), from an AD case (84 years) (arrow). (E, i^iii) A series of three optical sections taken through a cluster of DNs double labelled with antibodies to NF (green) and tau (red) in an AD case (84 years). The images were captured at intervals of approximately 2.8 Wm. The total depth of the sectioned cluster was 5.6 Wm. Images show numerous examples of NF-immunopositive bulb-like (arrowhead) and elongated (arrow) DNs with tau cores within the same plaque-associated DN cluster. Scale bar = 20 Wm (A and C), 15 Wm (B, D and E).
NSC 5014 16-7-01
Neuritic plaquesin Alzheimer's disease
exhibited considerable diversity with regard to the morphological types of L-amyloid deposits that were present. Although, historically, plaques have been grouped into varying numbers of di¡erent morphological types, this study found that all spherical plaques investigated could adequately be assigned to one of three distinct types, either dense-cored, ¢brillar or di¡use. Confocal microscopy was utilised to section through amyloid deposits to accurately categorise plaque type and to investigate the composition of individual plaques. Importantly, this procedure revealed a potential for incorrect plaque classi¢cation if only one focal plane was considered, as is often the case with routine microscopy. This was particularly important when discriminating between ¢brillar and dense-cored plaques, as well as for the identi¢cation of a corona of L-amyloid around dense-cored deposits that may have routinely been considered as `burnt-out' or `compact' plaques. The presence of L-amyloid plaques of varying morphology has been noted by a number of investigators (Wisniewski and Terry, 1973; Ulrich, 1985; Masliah et al., 1993a; Yasuhara et al., 1994; Schmidt et al., 1995; Dickson, 1997; Armstrong, 1998). Two main hypotheses have been developed to account for the observation. Firstly, that one type of plaque is converted into another, i.e. that the di¡erent plaque types represent stages in the life history of a single type of plaque (Ikeda et al., 1990). The second hypothesis is that each plaque type evolves independently of the others and, therefore, unique factors are involved in their formation (Armstrong, 1998). In this investigation the prevalence of di¡use, ¢brillar and dense-core plaques was variable between case types but did not correlate with disease progression as all plaque types were evident within both preclinical and end-stage AD. These data suggest that the `life history' theory of plaque development is unlikely to be strictly correct. Further to this, the presence of dense-cored plaques in preclinical AD cases suggests that these lesions are not con¢ned to cases in the very end stages of the disease (Ikeda et al., 1989; Rozemuller et al., 1989). Thus, the classi¢cation outlined using confocal microscopy in the current study allows the simpli¢cation of plaque typing, while avoiding terms that imply progressive maturation between plaque variants. All plaque types were present in both preclinical and end-stage AD cases, however, quantitative analysis demonstrated that there was a distinct shift in plaque pro¢le that correlated with disease progression. In the preclinical AD cases, a greater proportion of spherical plaques were diffuse in morphology, whereas, in AD cases, there were relatively more ¢brillar plaques and fewer di¡use spherical deposits. Interestingly, quantitative analysis con¢rmed that the proportion of dense-cored plaques was generally equivalent between these case types. A common phenomenon in AD is the association between plaques and clusters of DNs. As noted previously (Dickson et al., 1999), all clusters of abnormal neurites were found to be localised with L-amyloid deposits, however, not all deposits were associated with DNs. Analysis revealed that this association was not governed by the morphological characteristics of the
NSC 5014 16-7-01
105
plaque as all plaque types were associated with DN clusters, albeit to varying degrees. In addition, the rare but notable association of di¡use amyloid deposits with clusters of DNs, in both the preclinical and end-stage AD cases, further argues against the likelihood of di¡use amyloid representing an early, transient and inert stage of plaque development (Yamaguchi et al., 1988). However, it should be noted that in accordance with other investigators, dense-cored and ¢brillar plaques had a more pronounced e¡ect on surrounding neurites as compared to di¡use plaques (Knowles et al., 1999). In particular, ¢brillar plaques in end-stage AD demonstrated increased capacity for being `neuritic' compared to preclinical cases. These results suggest that certain plaque types are more deleterious to surrounding neuronal processes, however, further non-morphological traits may di¡erentiate damaging and non-damaging deposits. A notable morphological and neurochemical diversity has previously been reported in the abnormal neurites associated with L-amyloid deposits (Dickson et al., 1988, 1999; Vickers et al., 1994; Yasuhara et al., 1994; Wang and Munoz, 1995; Su et al., 1996; Saunders et al., 1998). Confocal microscopy provided a means to investigate the interrelationship between DNs and L-amyloid plaques in a manner not possible with a conventional microscope. Preclinical cases were characterised by the presence of NF-immunopositive bulb- and ring-like DNs that were not labelled with antibodies to tau, and end-stage cases, by NF- and tau-immunoreactive bulblike neurites and swollen fusiform processes (Dickson et al., 1999). The NF accumulations in the preclinical cases may represent one of the earliest pathological changes that occur within neurones in AD (Su et al., 1996; Vickers et al., 1996; Dickson et al., 1999). In both preclinical and end-stage AD cases, the bulb-like DNs occupied large pores within the plaque substructure. Cruz et al. (1997) also reported the occupation of such pores by cellular elements in AD cases. Similar pores were present in non-neuritic plaques. Therefore, it is unlikely that the physical presence of the abnormal neurite results in the formation of the pore. Alternatively, these abnormal neuronal processes may form in subregions of the plaque of least resistance. The more elongated NF-immunoreactive DNs are morphologically identical to those structures labelled by antibodies to tau in end-stage AD cases. These structures appeared as tortuous processes that tapered at the end most closely associated with the L-amyloid deposit. This taper may represent a point of focal constriction exerted by the plaque which initiates the accumulation of NFs within these processes. Other investigators have similarly described a swelling and disruption of normal morphology in axons (Benes et al., 1991) and dendrites (Knowles et al., 1999) observed passing through plaques. Optical sectioning revealed that the DNs observed in the current investigation were mostly non-continuous as opposed to single-branched processes, which has previously been proposed (Masliah et al., 1993a). Double labelling with antibodies to phosphorylated NFs (SMI312) and tau con¢rmed the presence of NFimmunoreactive bulb-like structures with a tau core
Cyaan Magenta Geel Zwart
106
T. C. Dickson and J. C. Vickers
which may correspond to a transitional stage of DN development (Dickson et al., 1999). Optical sectioning allowed for accurate quantitation of this class of abnormal neurite. Approximately 30% of NF-immunoreactive DNs in end-stage AD cases were found to have a tauimmunoreactive central region. Further to this, whereas previous reports describe this pattern of labelling in bulb-like DNs (Vickers et al., 1994; Dickson et al., 1999), it is now clear that this distinct pattern of labelling is also present in a proportion of classical elongated, tauimmunoreactive DNs. These results further implicate NFs in the development of not only the earliest abnormal neurites associated with plaque development but also the accumulation of tau within these processes.
both the early and late stages of AD and were capable of inducing local neuritic alterations. Quantitative analysis demonstrated that clinical AD was associated with a proportional shift to particular plaque types, principally the ¢brillar form, and that the ¢brillar plaques showed a higher tendency to be neuritic in end-stage AD cases. These data argue against a general toxicity of L-amyloid plaques. Rather, the ability of a plaque to cause neuronal damage was closely associated with its structural characteristics. The cytoskeletal involvement in DN formation also changes from preclinical to end-stage AD, and suggests that pathological changes in tau occur later in the disease process and are associated with clinical dementia (Thal et al., 1998).
CONCLUSION
AcknowledgementsöWe would like to acknowledge the assistance of the Sun Health Research Institute (Arizona, USA) in providing human brain material. This study was funded by the Department of Veterans A¡airs, The National Health and Medical Research Council and the Masonic Centenary Medical Research Foundation.
In this investigation, through the use of confocal imaging techniques, we de¢ned three major plaque types based on morphology. All plaque types were present in
REFERENCES
Armstrong, R.A., 1998. L-Amyloid plaques: Stages in life history or independent origin ? Dement. Geriatr. Cogn. Disord. 9, 227^238. Arnold, S.E., Hyman, B.T., Flory, J., Damasio, A.R., Van, H.G., 1991. The topographical and neuroanatomical distribution of neuro¢brillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer's disease. Cereb. Cortex 1, 103^116. Arriagada, P.V., Growdon, J.H., Hedly-Whyte, T., Hyman, B.T., 1992a. Neuro¢brillary tangles not senile plaques parallel duration and severity of Alzheimer's disease. Neurology 42, 631^639. Arriagada, P.V., Marzlo¡, K., Hyman, B.T., 1992b. Distribution of Alzheimer-type pathologic changes in nondemented elderly individuals matches the pattern in Alzheimer's disease. Neurology 42, 1681^1688. Benes, F.M., Farol, P.A., Majocha, R.E., Marotta, C.A., Bird, E.D., 1991. Evidence for axonal loss in regions occupied by senile plaques in Alzheimer cortex. Neuroscience 42, 651^660. Braak, H., Braak, E., 1991. Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol. (Berlin) 82, 239^259. Cruz, L., Urbanc, B., Buldyrev, S.V., Christie, R., Gomez-Isla, T., Havlin, S., McNamara, M., Stanley, H.E., Hyman, B.T., 1997. Aggregation and dissaggregation of senile plaques in Alzheimer's disease. Proc. Natl. Acad. Sci. USA 94, 7612^7616. Dickson, D.W., 1997. The pathogenesis of senile plaques. J. Neuropathol. Exp. Neurol. 56, 321^339. Dickson, D.W., Farlo, J., Davies, P., Crystal, H., Fuld, P., Yen, S.H., 1988. Alzheimer's disease : A double-labelling immunohistochemical study of senile plaques. Am. J. Pathol. 132, 86^101. Dickson, T.C., King, C.E., McCormack, G., Vickers, J.C., 1999. Neurochemical diversity of the dystrophic neurites in the early and late stages of Alzheimer's disease. Exp. Neurol. 156, 100^110. Hof, P.R., Morrison, J.H., 1994. The cellular basis of cortical disconnection in Alzheimer disease and related dementing conditions. In: Terry, R.D., Katzman, R., Bick, K.L. (Eds.), Alzheimer Disease. Raven, New York, pp. 197^229. Ikeda, S.I., Allsop, D., Glenner, G.G., 1989. Morphology and distribution of plaque and related deposits in the brains of Alzheimer's disease and control cases. An immunohistochemical study using amyloid L-protein antibody. Lab. Invest. 60, 113^122. Ikeda, S.I., Yanagisawa, M.B., Allsop, D., Glenner, G.G., 1990. Early senile plaques in Alzheimer's disease demonstrated by histochemistry, immunocytochemistry and electron microscopy. Hum. Pathol. 21, 1221^1226. Kato, S., Gondo, T., Hoshii, Y., Takahashi, M., Yamada, M., Ishihara, T., 1998. Confocal observation of senile plaques in Alzheimer's disease : Senile plaque morphology and relationship between senile plaques and astrocytes. Pathol. Int. 48, 332^340. Knowles, R.B., Wyart, C., Buldyrev, S.V., Cruz, L., Urbanc, B., Hasselmo, M.E., Stanley, H.E., Hyman, B.T., 1999. Plaque-induced neurite abnormalities : Implications for disruption of neural networks in Alzheimer's disease. Proc. Natl. Acad. Sci. USA 96, 5274^5279. Masliah, E., Mallory, M., Deernink, T., DeTeresa, R., Lamont, S., Miller, A., Terry, R.D., Carragher, R., Elisman, M., 1993a. Re-evaluation of the structural organization of neuritic plaques in Alzheimer's disease. J. Neuropathol. Exp. Neurol. 52, 619^632. Masliah, E., Mallory, M., Hansen, L., Alford, M., DeTeresa, R., Terry, R.D., 1993b. An antibody against phosphorylated neuro¢laments identi¢es a subset of damaged association axons in Alzheimer's disease. Am. J. Pathol. 142, 871^882. Mirra, S.S., Heyman, A., McKeel, D., Sumi, S.M., Crain, B.J., Brownlee, L.M., Vogel, F.S., Hughes, J.P., Van Belle, B., Berg, L., 1991. The consortium to establish a registry for Alzheimer's disease (CERAD). Part II. Standardization of the neuropathological assessment of Alzheimer's disease. Neurology 41, 479^486. Morris, J.C., McKeel, D.W.Jr., Storandt, M., Rubin, E.H., Price, J.L., Grant, E.A., Ball, M.J., Berg, L., 1991. Very mild Alzheimer's disease : informant-based clinical, psychometric, and pathologic distinction from normal aging. Neurology 41, 469^478. Morris, J.C., Storandt, M., McKeel, D.W., Rubin, E.H., Price, J.L., Grant, E.A., Berg, L., 1996. Cerebral amyloid deposition and di¡use plaques in `normal' aging: Evidence from presymptomatic and very mild Alzheimer's disease. Neurology 46, 707^719. Price, J.L., Morris, J.C., 1999. Tangles and plaques in nondemented aging and `preclinical' Alzheimer's disease. Ann. Neurol. 45, 358^368. Rozemuller, J.M., Eikelenboom, P., Stam, F.C., Beyreuther, K., Masters, C., 1989. A4 protein in Alzheimer's disease: Primary and secondary cellular events in extracellular amyloid deposition. J. Neuropathol. Exp. Neurol. 48, 674^691. Saunders, H.L., Dickson, T.C., Vickers, J.C., 1998. Immunolocalisation of L-amyloid precursor protein to plaque-associated synaptic alterations in the early and late stages of Alzheimer's disease. Alzheimer's Rep. 1, 111^119.
NSC 5014 16-7-01
Neuritic plaquesin Alzheimer's disease
107
Schmidt, M.L., Robinson, K.A., Lee, V.M-Y., Trojanowski, J.Q., 1995. Chemical and immunological heterogeneity of ¢brillar amyloid in plaques of Alzheimer's disease and Down's syndrome brains revealed by confocal microscopy. Am. J. Pathol. 147, 503^515. Selkoe, D.J., 1994. Alzheimer's disease: a central role for amyloid. J. Neuropathol. Exp. Neurol. 53, 438^447. Su, J.H., Cummings, B.J., Cotman, C.W., 1996. Plaque biogenesis in brain aging and Alzheimer's disease. I. Progressive changes in phosphorylation states of paired helical ¢laments and neuro¢laments. Brain Res. 739, 79^87. Terry, R.D., 1996. The pathogenesis of Alzheimer disease : An alternative to the amyloid hypothesis. J. Neuropathol. Exp. Neurol. 55, 1023^1025. Thal, D.R., Arendt, T., Waldmann, G., Holzer, M., Zedlick, D., Rub, U., Schober, R., 1998. Progression of neuro¢brillary changes and PHF-d in end-stage Alzheimer's disease is di¡erent from plaque and cortical microglial pathology. Neurobiol. Aging 19, 517^525. Ulrich, J., 1985. Alzheimer changes in nondemented patients younger than sixty-¢ve: Possible early stages of Alzheimer's disease and senile dementia of Alzheimer type. Ann. Neurol. 17, 273^277. Vickers, J.C., Chin, D., Edwards, A.-M., Sampson, V., Harper, C., Morrison, J., 1996. Dystrophic neurite formation associates with age-related L amyloid deposition in the neocortex : clues to the genesis of neuro¢brillary pathology. Exp. Neurol. 141, 1^11. Vickers, J.C., Dickson, T.C., Adlard, P.A., Saunders, H.L., King, C.E., McCormack, G., 2000. The cause of neuronal degeneration in Alzheimer's disease. Prog. Neurobiol. 60, 139^165. Vickers, J.C., Riederer, B.M., Marugg, R.A., Buee-Scherrer, V., Buee, L., Delacourte, A., Morrison, J.H., 1994. Alterations in neuro¢lament protein immunoreactivity in human hippocampal neurons related to normal aging and Alzheimer's disease. Neuroscience 62, 1^13. Wang, D., Munoz, D.G., 1995. Qualitative and quantitative di¡erences in senile plaque dystrophic neurites of Alzheimer's disease and normal aged brain. J. Neuropathol. Exp. Neurol. 54, 584^586. Wisniewski, H.M., Bancher, C., Barcikowska, M., Wen, G.Y., Currie, J., 1989. Spectrum of morphological appearance of amyloid deposits in Alzheimer's disease. Acta Neuropathol. 78, 337^347. Wisniewski, H.M., Terry, R.D., 1973. Reeaxamination of the pathogenesis of the senile plaque. Prog. Neuropathol. 2, 1^26. Yamaguchi, H., Hirai, S., Morimatsu, M., Shoji, M., Harigaya, Y., 1988. Di¡use type of senile plaques in the brains of Alzheimer-type dementia. Acta Neuropathol. 77, 113^119. Yasuhara, O., Kawamata, T., Aimi, Y., McGeer, E.G., McGeer, P.L., 1994. Two types of dystrophic neurites in senile plaques of Alzheimer disease and elderly non-demented cases. Neurosci. Lett. 171, 73^76. (Accepted 3 April 2001)
NSC 5014 16-7-01
Cyaan Magenta Geel Zwart
Neuroscience Vol. 105, No. 1, pp. 99^107, 2001 ß 2001 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306-4522 / 01 $20.00+0.00
www.elsevier.com/locate/neuroscience
THE MORPHOLOGICAL PHENOTYPE OF L-AMYLOID PLAQUES AND ASSOCIATED NEURITIC CHANGES IN ALZHEIMER'S DISEASE T. C. DICKSON and J. C. VICKERS* Discipline of Pathology, Clinical School, University of Tasmania, 43 Collins St, Hobart, Tasmania 7000, Australia
AbstractöWe have utilised laser confocal microscopy to categorise L-amyloid plaque types that are associated with preclinical and end-stage Alzheimer's disease and to de¢ne the neurochemistry of dystrophic neurites associated with various forms of plaques. Plaques with a spherical pro¢le were de¢ned as either di¡use, ¢brillar or dense-cored using Thio£avin S staining or immunolabelling for L-amyloid. Confocal analysis demonstrated that ¢brillar plaques had a central mass of L-amyloid with compact spoke-like extensions leading to a con£uent outer rim. Dense-cored plaques had a compacted central mass surrounded by an outer sphere of L-amyloid. Di¡use plaques lacked a morphologically identi¢able substructure, resembling a ball of homogeneous labelling. The relative proportion of di¡use, ¢brillar and dense-cored plaques was 53, 22 and 25% in preclinical and 31, 49 and 20% in end-stage Alzheimer's disease cases, respectively. Plaque-associated dystrophic neurites in preclinical cases were immunolabelled for neuro¢lament proteins whereas, in end-stage cases, these abnormal neurites were variably labelled for tau and/or neuro¢laments. Double labelling demonstrated that the proportion of di¡use, ¢brillar and dense-cored plaques that were neuritic was 12, 47 and 82% and 24, 82 and 76% in preclinical and end-stage cases, respectively. Most dystrophic neurites in Alzheimer's disease cases were labelled for either neuro¢laments or tau, however, confocal analysis determined that 30% of neuro¢lament-labelled bulb-like or elongated neurites had a core of tau immunoreactivity. These results indicate that all morphologically de¢ned L-amyloid plaque variants were present in both early and late stages of Alzheimer's disease. However, progression to clinical dementia was associated with both a shift to a higher proportion of ¢brillar plaques that induced local neuritic alterations and a transformation of cytoskeletal proteins within associated abnormal neuronal processes. There data indicate key pathological changes that may be subject to therapeutic intervention to slow the progression of Alzheimer's disease. ß 2001 IBRO. Published by Elsevier Science Ltd. All rights reserved. Key words: Alzheimer's disease, L-Amyloid, plaque, dystrophic neurites, neuro¢laments, tau.
may be a key mechanism in the development of AD, the presence of L-amyloid-containing plaques may represent presymptomatic or unrecognised early symptomatic AD (Morris et al., 1991, 1996; Price and Morris, 1999). In contrast, the relative density of plaques may decrease with advanced stages of the disease, presumably due to clearance by associated glial cell types (Thal et al., 1998). The heterogeneity of plaque morphology has also been noted by a number of early investigators in the ¢eld of AD research (Wisniewski and Terry, 1973; Ulrich, 1985), but the pathological signi¢cance of these plaque isoforms to the disease mechanism is yet to be de¢nitively determined (Armstrong, 1998; Ikeda et al., 1990). This variable plaque morphology has led to the development of a number of di¡erent typing schemes that have been consistently modi¢ed and updated with the advent of more sensitive staining and immunolabelling methods. The form of microscopy used in individual investigations has also been recognised as an important contributor to the ability to accurately describe plaque morphology. The localisation of clusters of abnormal neuronal processes, referred to as dystrophic neurites (DNs), with a subset of the L-amyloid plaques is a common phenomenon in AD (Hof and Morrison, 1994). These neuritic plaques have been considered a pathological correlate of AD dementia (Mirra et al., 1991). Standard immunohis-
The deposition of L-amyloid in the brain is a generally accepted, but not speci¢c, pathological hallmark of Alzheimer's disease (AD), however, there is considerable diversity of opinion regarding the signi¢cance of plaque formation to the aetiology of the disease (Arnold et al., 1991; Arriagada et al., 1992a; Selkoe, 1994; Terry, 1996). For example, there have been inconsistent reports on the correlation between L-amyloid plaque density and various measures of clinical de¢cit, and there is an alternative view that more soluble L-amyloid may be of greater pathological importance than the insoluble L-amyloid that comprises plaques (Vickers et al., 2000). Furthermore, in elderly persons without overt dementia, the presence of plaques in the neocortex is often accepted as part of normal aging (Arriagada et al., 1992b), therefore making the determination of the pathological significance of such structures di¤cult. Other investigators suggest that because cerebral deposition of L-amyloid
*Corresponding author. Tel. : +61-3-62264830; fax: +61-362264833. E-mail address: [email protected] (J. C. Vickers). Abbreviations : AD, Alzheimer's disease ; DN, dystrophic neurite; NF, neuro¢lament ; PBS, phosphate-bu¡ered saline; SFG, superior frontal gyrus. 99
NSC 5014 16-7-01
Cyaan Magenta Geel Zwart
100
T. C. Dickson and J. C. Vickers
tochemical methods and £uorescence microscopy have been previously utilised to e¡ectively analyse the exact complement of epitopes expressed within the clusters of abnormal neurites associated with L-amyloid plaques (Dickson et al., 1988, 1999; Masliah et al., 1993b; Vickers et al., 1994, 1996; Yasuhara et al., 1994; Wang and Munoz, 1995; Su et al., 1996; Saunders et al., 1998; Thal et al., 1998;). Confocal microscopy allows for the detailed study of this association between L-amyloid plaques and DNs. Masliah et al. (1993a), through the use of confocal microscopy, concluded, in their investigation of the alterations that occur in synapses and axons in AD, that many of the tau abnormal neurites commonly localised with L-amyloid plaques were continuous with synaptophysin-positive distended terminals. Others have used confocal microscopy to investigate various aspects of senile plaque morphology in end-stage AD cases (Schmidt et al., 1995; Cruz et al., 1997) and also the relationship between senile plaques and astrocytes (Kato et al., 1998). In the current investigation, confocal microscopy was utilised to accurately quantitate the proportion of various plaque morphological types in both preclinical and end-stage AD cases and to investigate their association with neurochemically de¢ned clusters of DNs. These data provide new information on the link between two of the hallmarks of AD, the L-amyloid plaque and DNs, and further elucidate the sequence of pathology development that ultimately leads to nerve cell degeneration and dementia.
EXPERIMENTAL PROCEDURES
Tissue source and processing Blocks of human brain tissue were obtained from the Sun Health Research Institute (Arizona, USA) as previously described (Saunders et al., 1998). Informed consent for the collection of material was obtained prior to death and tissue use was approved by the Institutional Ethics Committee and is consistent with the Declaration of Helsinki. These blocks of cerebral cortex (superior frontal gyrus (SFG), Brodmann area 9) were immersion-¢xed in 4% paraformaldehyde, cryoprotected and sectioned on a freezing microtome at 40^50 Wm. These samples included ¢ve cases of AD conforming to the CERAD (Consortium to Establish a Registry for Alzheimer's Disease) criteria (Mirra et al., 1991) (74, 74, 83, 88 and 92 years of age), and ¢ve preclinical AD cases (78, 81, 84, 90 and 91 years of age). The post-mortem interval for all cases ranged from 2 to 3 h. No di¡erences were observed in the immunolabelling pro¢le for the various antibodies across varying postmortem to ¢xation intervals or di¡erent ¢xation protocols. The preclinical cases were de¢ned by the lack of overt dementia (although de¢nitive neuropsychological data were not available)
and the presence of widespread L-amyloid immunoreactive plaques in the neocortex, with the distinction that these plaques could not be de¢ned as `neuritic' based on Thio£avin S staining or immunolabelling for PHF-tau or ubiquitin (Vickers et al., 1996). Thus, these preclinical cases do not conform to CERAD criteria for diagnosis of clinical AD (Mirra et al., 1991). These cases do correspond to Braak staging of `3' based on the presence of neuro¢brillary tangles in the entorhinal cortex and hippocampus as well as L-amyloid plaques in the neocortex (Braak and Braak, 1991). Notably, these preclinical cases lacked neocortical neuro¢brillary tangles. Staining To investigate the morphology of L-amyloid plaques, 100-Wm sections of SFG from the preclinical and end-stage AD cases were stained with Thio£avin S [0.0125% in 40% ethanol and 60% 0.01 M phosphate-bu¡ered saline (PBS)] (Sigma, St. Louis, MO, USA) for 3 min at room temperature, protected from the light. After staining, sections were di¡erentiated in two 10-min rinses of 50% ethanol and 50% 0.01 M PBS before ¢nal washes in 0.01 M PBS. Immunohistochemistry The relationship between neuro¢lament (NF)-immunoreactive DNs and L-amyloid was investigated in all cases. Sections in which L-amyloid was to be visualised were pre-treated in 90% formic acid prior to incubation in either primary antibody. A rabbit antibody which recognises all carboxy-terminal variants of L-amyloid (AL1^40 and AL1^42(43)), pan L-amyloid, was combined alternatively with mouse monoclonal antibodies to phosphorylated NFs (SMI312) in both preclinical and endstage AD cases. The co-localisation pattern of cytoskeletal proteins (NF, tau) within individual abnormal neurites was also investigated using double labelling immunohistochemistry by combining the rabbit antibody to 4Rtau (four microtubule binding domains, raised against a non-phosphorylated recombinant protein) with the SMI312 mouse monoclonal NF antibody (see Table 1 for primary antibody details). SMI312 is a `cocktail' of monoclonal antibodies to phosphorylated NFs (middle and high molecular weight subunits) and does not cross-react with tau in AD cases (Dickson et al., 1999). For all £uorescence labelling, mouse monoclonal antibodies were visualised with a horse antimouse secondary antibody conjugated to £uorescein isothiocyanate (FITC; Vector Laboratories, Burlingame, CA, USA; 1:200) whereas the rabbit polyclonal was visualised with a goat anti-rabbit antibody conjugated to biotin (Vector, 1:200) followed by Texas Red avidin D (Vector, 1:200). In all cases, the density (/mm2 ) of L-amyloid labelled plaques was quantitated. A rectangular counting frame and 10U objective were utilised to count spherical L-amyloid-immunoreactive deposits (greater than 20 Wm diameter) in the neocortex from non-overlapping ¢elds, representing ¢ve pia to white matter traverses, examined with a Leitz epi£uorescence microscope. Confocal microscopy Laser confocal scanning microscopy, using an Optiscan F900e krypton/argon system attached to an Olympus BX50 epi£uorescence microscope, was utilised to investigate the morphology of various plaque types and the localisation patterns of DNs
Table 1. Antibody details Code
Type
Reactivity
Dilution
Source
SMI32 SMI312 L-Amyloid (pan) Tau
m mc r r
dephosphorylated NF-M and NF-H phosphorylated NF-M and NF-H All L-amyloid peptides 4R Tau
1:1000 1:1000 1:500 1:2000
Sternberger Monoclonals Inc., Lutherville, USA Sternberger Monoclonals Inc., Lutherville, USA QCB, Hopkinton, USA Dako, Glostrup, Denmark
m: mouse monoclonal; mc: mouse monoclonal cocktail; r: rabbit polyclonal.
NSC 5014 16-7-01
Neuritic plaquesin Alzheimer's disease
within the plaque substructure. Thio£avin S-stained and Texas Red-labelled preparations were recorded simultaneously in separate channels. Representative plaques were optically sectioned at intervals of 1^2 Wm for distances of approximately 20^30 Wm through a plaque. This information was then utilised to create three-dimensional images of the varying plaque types present throughout the cases. Plaques were categorised as either `spherical-di¡use', `dense-cored' or `¢brillar'. Similar sectioning methods were utilised for material double labelled with the antibodies to L-amyloid, however, distances of only 5^10 Wm were sectioned due to antibody penetration limitations. Detailed analysis of the plaque substructure was performed using the Optimate 5.2 program. The relative proportion of each plaque type was determined by examining in detail 100 plaques, throughout all cortical layers, in each case (i.e. 1000 plaques in total). Further to this, the proportion of each of these plaque types that were neuritic, i.e. associated with either NF-labelled (preclinical AD cases) (Dickson et al., 1999) or tau-labelled (end-stage AD cases) abnormal neurites, was also determined in each case. Di¡use amyloid was noted to be highly variable in its size and deposition patterns. Therefore, for the purpose of these investigations, only spherical di¡use deposits with a minimum diameter of approximately 40 Wm were considered. Confocal microscopy was further utilised to perform an analysis of the plaque-associated DNs in the AD cases. The 40-Wm sections double labelled with the rabbit antibody to tau and the mouse antibody to phosphorylated (SMI312) NFs were used for these analyses. Clusters of DNs were optically sectioned at 0.5^ 1 Wm intervals. `Optimate' was again utilised for size measurements. The proportion of NF-immunoreactive DNs that had a tau-immunoreactive core was quanti¢ed by sectioning through 100 non-continuous DNs labelled by SMI312 in each AD case. These counts were made throughout the layers, and included all morphological types of DNs.
RESULTS
Preclinical Alzheimer's disease Staining with Thio£avin S and labelling with antibodies to L-amyloid revealed a number of morphologically distinct plaques. Serial `confocal' sectioning throughout these plaques uncovered key features speci¢c to each plaque type. A subset of plaques had a dense core of L-amyloid, typically surrounded by a ring of indistinct wispy labelling (Fig. 1A). Other plaques lacked a dense central core and were comprised of large numbers of distinct, spoke-like ¢bril bundles that emanate from a central mass. These `¢brillar' plaques had distinct pores
101
and irregularities within their structure (Fig. 1B). Of note, the L-amyloid labelling in the outer regions of both of these plaque types was usually con£uent. Di¡use spherical deposits were more irregular in shape and without distinct edges (Fig. 1C). Layer I of the neocortex was notable for its lack of any form of L-amyloid labelling in preclinical AD cases. Variation between cases was low with regard to patterns of L-amyloid labelling. Quantitation of the proportion of these three plaque types in ¢ve cases demonstrated that approximately half of the spherical L-amyloid deposits were di¡use (average of 53.4% þ 2.2, S.E.M.), with dense-cored and ¢brillar plaques accounting for the remaining 24.6% ( þ 2.2, S.E.M.) and 22.0% ( þ 1.7, S.E.M.) of plaques, respectively (Table 2). The mean proportion of the plaques in these cases that were neuritic, as de¢ned by the co-presence of NFlabelled DNs, was 36.8% ( þ 7.2, S.E.M.). Of the di¡use plaques, 12% ( þ 2.1, S.E.M.) were associated with NFimmunopositive DNs. However, 47.0% ( þ 3.6, S.E.M.) of ¢brillar plaques and 81.8% ( þ 3.4, S.E.M.) of densecored plaques were neuritic (Table 3). The NF-immunoreactive DNs exhibited a variety of morphological types including bulb- and ring-like structures. These structures were labelled by SMI312 but not labelled by the antibody to tau. Optical sectioning through neuritic plaques indicated that the larger bulb-like neurites were generally located within larger pores in the amyloid plaque substructure (Fig. 2A). In the dense-cored plaques, these pores mostly occupied the area directly adjacent to the L-amyloid core, whereas, in di¡use and ¢brillar plaques, they were non-uniformly dispersed throughout the plaque. The larger bulb-like DNs were often continuous with ¢ne processes of normal diameter. Many of these processes appeared to pass through pores in the plaque, immediately terminating with a large swelling (Fig. 2A). The ring-like structures also appeared to be co-located to spaces within the L-amyloid plaque, but these internal subregions devoid of amyloid were smaller in dimension to those occupied by the larger bulbs. Alzheimer's disease The same morphological variants of spherical L-amy-
Table 2. L-Amyloid plaque variants
Preclinical case 1 2 3 4 5 Mean ( þ S.E.M.) End-stage case 1 2 3 4 5 Mean ( þ S.E.M.)
Plaque density (/mm2 )
% Fibrillar
% Dense-cored
% Di¡use
21.9 31.2 10.3 10.3 15.4 17.8 þ 4.0
27 17 22 24 20 22.0 þ 1.7
26 23 23 25 26 24.6 þ 0.7
47 60 55 51 54 53.4 þ 2.2
23.0 42.9 40.8 47.5 48.6 40.6 þ 4.6
58 56 43 50 40 49.4 þ 3.5
22 8 22 19 28 19.8 þ 3.3
20 36 35 31 32 30.8 þ 2.9
NSC 5014 16-7-01
Cyaan Magenta Geel Zwart
102
T. C. Dickson and J. C. Vickers
Fig. 1. (A, i^iii) Series of three optical sections taken through a dense-cored plaque stained with Thio£avin S in a preclinical AD case (83 years). The images were captured at intervals of approximately 14.5 Wm. The total depth of the sectioned plaque was 28.56 Wm. Images show a distinct central core (arrow) of L-amyloid surrounded by a void or clearing and then an outer spherical rim of L-amyloid (arrowhead). (B, i^iii) A similar set of images sectioned through a ¢brillar plaque stained with Thio£avin S in a preclinical AD case (81 years). The images were captured at intervals of approximately 17.5 Wm. The total depth of the sectioned plaque was 34.68 Wm. The ¢brillar plaques showed dense L-amyloid accumulations throughout the plaque structure. In the central regions these accumulations often appeared as spoke-like accumulations (arrow), emanating from a denser central accumulation. Distinct pores within the plaque were obvious in deeper sections. (C, i^iii) A set of sequential confocal images sectioned through a di¡use plaque labelled with an antibody to L-amyloid in an end-stage AD case (74 years). The images were captured at intervals of approximately 2.5 Wm. The total depth of the plaque that was sectioned was 5 Wm. These deposits were generally larger in diameter, however, considerably smaller in depth. Only small changes in the plaque structure were noted in the di¡erent optical sections. Scale bar = 20 Wm (A and B), 50 Wm (C).
loid deposits were present in the end-stage AD cases as those noted in the preclinical material, i.e. di¡use, ¢brillar and dense-cored. L-Amyloid labelling was, however, more intense and spread throughout the cortical layers. In addition, there were approximately twice as many L-amyloid plaques in the neocortex of end-stage cases
relative to preclinical cases (Table 3), and this di¡erence was statistically signi¢cant (P 6 0.01). In end-stage AD, large numbers of di¡use L-amyloid-immunopositive deposits were present throughout layer I with many dense-cored plaques present within layer V. Interestingly, whereas other studies have noted a morphological form
NSC 5014 16-7-01
Neuritic plaquesin Alzheimer's disease
103
Fig. 2 (Caption overleaf).
of L-amyloid deposit consisting solely of a dense central deposit (so-called `compact' or `burnt-out' plaques, reviewed by Armstrong (1998)), confocal analysis in this study determined that these were always associated
NSC 5014 16-7-01
with some degree of a corona of surrounding L-amyloid immunoreactivity, thus conforming to the dense-cored classi¢cation. The prevalence of the three spherical plaque types was notably di¡erent in the AD cases as
Cyaan Magenta Geel Zwart
104
T. C. Dickson and J. C. Vickers Table 3. Proportion (%) of L-amyloid plaque variants associated with dystrophic neurites
Preclinical case 1 2 3 4 5 Mean ( þ S.E.M.) End-stage case 1 2 3 4 5 Mean ( þ S.E.M.)
Fibrillar
Dense-cored
Di¡use
57 38 47 40 53 47.0 þ 3.6
92 73 86 82 76 81.8 þ 3.4
17 6 8 14 15 12.0 þ 2.1
83 86 77 82 83 82.2 þ 1.5
81 75 73 74 79 76.4 þ 1.5
20 28 22 32 19 24.2 þ 2.5
compared to the preclinical AD cases. In ¢ve AD cases, proportional analysis demonstrated that 49.4% ( þ 3.5, S.E.M.) of spherical L-amyloid deposits were ¢brillar, 19.8% ( þ 3.3, S.E.M.) dense-cored and 30.8% ( þ 2.9, S.E.M.) di¡use (Table 2). Analysis of variance demonstrated a signi¢cant interaction e¡ect (P 6 0.01) when comparing the proportion of plaque types in preclinical vs end-stage AD cases, which can be attributed to the shift to a higher relative proportion of di¡use to ¢brillar plaque types from preclinical to end-stage AD cases. In this respect, there was a signi¢cant di¡erence (P 6 0.01) in the mean proportion of di¡use and ¢brillar plaques between the case categories, but interestingly, no signi¢cant di¡erence in the proportion of dense-cored plaques between preclinical and end-stage AD cases. Labelling of Thio£avin S-stained sections with an antibody to tau demonstrated that 76.4% ( þ 1.5, S.E.M.) of dense-cored plaques, 82.2% ( þ 1.5, S.E.M.) of ¢brillar plaques and 24.2% ( þ 2.5, S.E.M.) of di¡use deposits were neuritic (Table 3). The mean proportion of all plaques that were neuritic was 64.0% ( þ 3.3, S.E.M.). The relatively higher proportion of both di¡use and ¢brillar plaques associated with abnormal neurites in end-stage relative to preclinical AD cases was con¢rmed by statistical analysis (P 6 0.01). However, there was no signi¢cant di¡erence in the proportion of neuritic dense-cored plaques in either case type, where the relative proportion of this plaque types associated with DNs was high. The predominant morphological DN types included bulb-like structures of varying sizes and also elongated, fusiform structures (Fig. 2). Both forms exhibited immunoreactivity to anti-tau and anti-NF antibodies. The pat-
tern of distribution of the bulb-like DNs closely resembled that encountered for similar DNs in the preclinical cases. Generally, the bulb-like structures were located within the pores of the plaque (Fig. 2B). The more tortuous and elongated DNs had a somewhat different distribution within the L-amyloid plaques. Rather than occupying pores, these abnormal neurites emerged from small pores and did not appear to be associated with structural features of the surrounding L-amyloid. Consequently, they were located in all areas of plaques including within the central region (Fig. 2B). Analysis of the NF-immunopositive DNs by confocal microscopy con¢rmed the presence of a tau-immunoreactive core within a subset of these pathological structures. Initially this pattern of labelling was thought to be restricted to the bulb-like DNs (Fig. 2C) (Dickson et al., 1999), however, confocal microscopy also revealed tauimmunoreactive cores within the longer, more cylindrical DNs (Fig. 2D). In these structures only the very outer rim showed anti-NF labelling. Confocal investigation of individual NF-immunopositive DNs demonstrated that an average of 30.0% ( þ 4.3, S.E.M.) had a tau core. Many of these cores were only visible after optically sectioning through the DN clusters (Fig. 2E).
DISCUSSION
The current investigation con¢rms previous reports that L-amyloid plaques are not uniform structures (Ikeda et al., 1989; Wisniewski et al., 1989; Dickson, 1997). Both the preclinical and end-stage AD cases
Fig. 2. Double labelling with antibodies to L-amyloid (red) and phosphorylated NFs (green) showing the relationship between plaque morphology and the location of the abnormal neurites in a preclinical AD case (91 years) (A). Arrowheads indicate neuro¢lamentous ring-like structures and arrows, bulb-like structures. (B) Double labelling with antibodies to L-amyloid (red) and NF (green) showing the relationship between plaque morphology and the location of the abnormal neurites in an endstage AD case (92 years) Arrowheads indicate long distended DNs exhibiting classical tau morphology, whereas arrows indicate neuro¢lamentous bulb-like accumulations of similar morphology to those noted in preclinical AD cases. (C) Double labelling of bulb-like NF-immunopositive DNs (green) with a tau core (red) in an AD case (74 years). (D) A similar labelling pattern in an elongated fusiform NF-immunopositive DN (green), from an AD case (84 years) (arrow). (E, i^iii) A series of three optical sections taken through a cluster of DNs double labelled with antibodies to NF (green) and tau (red) in an AD case (84 years). The images were captured at intervals of approximately 2.8 Wm. The total depth of the sectioned cluster was 5.6 Wm. Images show numerous examples of NF-immunopositive bulb-like (arrowhead) and elongated (arrow) DNs with tau cores within the same plaque-associated DN cluster. Scale bar = 20 Wm (A and C), 15 Wm (B, D and E).
NSC 5014 16-7-01
Neuritic plaquesin Alzheimer's disease
exhibited considerable diversity with regard to the morphological types of L-amyloid deposits that were present. Although, historically, plaques have been grouped into varying numbers of di¡erent morphological types, this study found that all spherical plaques investigated could adequately be assigned to one of three distinct types, either dense-cored, ¢brillar or di¡use. Confocal microscopy was utilised to section through amyloid deposits to accurately categorise plaque type and to investigate the composition of individual plaques. Importantly, this procedure revealed a potential for incorrect plaque classi¢cation if only one focal plane was considered, as is often the case with routine microscopy. This was particularly important when discriminating between ¢brillar and dense-cored plaques, as well as for the identi¢cation of a corona of L-amyloid around dense-cored deposits that may have routinely been considered as `burnt-out' or `compact' plaques. The presence of L-amyloid plaques of varying morphology has been noted by a number of investigators (Wisniewski and Terry, 1973; Ulrich, 1985; Masliah et al., 1993a; Yasuhara et al., 1994; Schmidt et al., 1995; Dickson, 1997; Armstrong, 1998). Two main hypotheses have been developed to account for the observation. Firstly, that one type of plaque is converted into another, i.e. that the di¡erent plaque types represent stages in the life history of a single type of plaque (Ikeda et al., 1990). The second hypothesis is that each plaque type evolves independently of the others and, therefore, unique factors are involved in their formation (Armstrong, 1998). In this investigation the prevalence of di¡use, ¢brillar and dense-core plaques was variable between case types but did not correlate with disease progression as all plaque types were evident within both preclinical and end-stage AD. These data suggest that the `life history' theory of plaque development is unlikely to be strictly correct. Further to this, the presence of dense-cored plaques in preclinical AD cases suggests that these lesions are not con¢ned to cases in the very end stages of the disease (Ikeda et al., 1989; Rozemuller et al., 1989). Thus, the classi¢cation outlined using confocal microscopy in the current study allows the simpli¢cation of plaque typing, while avoiding terms that imply progressive maturation between plaque variants. All plaque types were present in both preclinical and end-stage AD cases, however, quantitative analysis demonstrated that there was a distinct shift in plaque pro¢le that correlated with disease progression. In the preclinical AD cases, a greater proportion of spherical plaques were diffuse in morphology, whereas, in AD cases, there were relatively more ¢brillar plaques and fewer di¡use spherical deposits. Interestingly, quantitative analysis con¢rmed that the proportion of dense-cored plaques was generally equivalent between these case types. A common phenomenon in AD is the association between plaques and clusters of DNs. As noted previously (Dickson et al., 1999), all clusters of abnormal neurites were found to be localised with L-amyloid deposits, however, not all deposits were associated with DNs. Analysis revealed that this association was not governed by the morphological characteristics of the
NSC 5014 16-7-01
105
plaque as all plaque types were associated with DN clusters, albeit to varying degrees. In addition, the rare but notable association of di¡use amyloid deposits with clusters of DNs, in both the preclinical and end-stage AD cases, further argues against the likelihood of di¡use amyloid representing an early, transient and inert stage of plaque development (Yamaguchi et al., 1988). However, it should be noted that in accordance with other investigators, dense-cored and ¢brillar plaques had a more pronounced e¡ect on surrounding neurites as compared to di¡use plaques (Knowles et al., 1999). In particular, ¢brillar plaques in end-stage AD demonstrated increased capacity for being `neuritic' compared to preclinical cases. These results suggest that certain plaque types are more deleterious to surrounding neuronal processes, however, further non-morphological traits may di¡erentiate damaging and non-damaging deposits. A notable morphological and neurochemical diversity has previously been reported in the abnormal neurites associated with L-amyloid deposits (Dickson et al., 1988, 1999; Vickers et al., 1994; Yasuhara et al., 1994; Wang and Munoz, 1995; Su et al., 1996; Saunders et al., 1998). Confocal microscopy provided a means to investigate the interrelationship between DNs and L-amyloid plaques in a manner not possible with a conventional microscope. Preclinical cases were characterised by the presence of NF-immunopositive bulb- and ring-like DNs that were not labelled with antibodies to tau, and end-stage cases, by NF- and tau-immunoreactive bulblike neurites and swollen fusiform processes (Dickson et al., 1999). The NF accumulations in the preclinical cases may represent one of the earliest pathological changes that occur within neurones in AD (Su et al., 1996; Vickers et al., 1996; Dickson et al., 1999). In both preclinical and end-stage AD cases, the bulb-like DNs occupied large pores within the plaque substructure. Cruz et al. (1997) also reported the occupation of such pores by cellular elements in AD cases. Similar pores were present in non-neuritic plaques. Therefore, it is unlikely that the physical presence of the abnormal neurite results in the formation of the pore. Alternatively, these abnormal neuronal processes may form in subregions of the plaque of least resistance. The more elongated NF-immunoreactive DNs are morphologically identical to those structures labelled by antibodies to tau in end-stage AD cases. These structures appeared as tortuous processes that tapered at the end most closely associated with the L-amyloid deposit. This taper may represent a point of focal constriction exerted by the plaque which initiates the accumulation of NFs within these processes. Other investigators have similarly described a swelling and disruption of normal morphology in axons (Benes et al., 1991) and dendrites (Knowles et al., 1999) observed passing through plaques. Optical sectioning revealed that the DNs observed in the current investigation were mostly non-continuous as opposed to single-branched processes, which has previously been proposed (Masliah et al., 1993a). Double labelling with antibodies to phosphorylated NFs (SMI312) and tau con¢rmed the presence of NFimmunoreactive bulb-like structures with a tau core
Cyaan Magenta Geel Zwart
106
T. C. Dickson and J. C. Vickers
which may correspond to a transitional stage of DN development (Dickson et al., 1999). Optical sectioning allowed for accurate quantitation of this class of abnormal neurite. Approximately 30% of NF-immunoreactive DNs in end-stage AD cases were found to have a tauimmunoreactive central region. Further to this, whereas previous reports describe this pattern of labelling in bulb-like DNs (Vickers et al., 1994; Dickson et al., 1999), it is now clear that this distinct pattern of labelling is also present in a proportion of classical elongated, tauimmunoreactive DNs. These results further implicate NFs in the development of not only the earliest abnormal neurites associated with plaque development but also the accumulation of tau within these processes.
both the early and late stages of AD and were capable of inducing local neuritic alterations. Quantitative analysis demonstrated that clinical AD was associated with a proportional shift to particular plaque types, principally the ¢brillar form, and that the ¢brillar plaques showed a higher tendency to be neuritic in end-stage AD cases. These data argue against a general toxicity of L-amyloid plaques. Rather, the ability of a plaque to cause neuronal damage was closely associated with its structural characteristics. The cytoskeletal involvement in DN formation also changes from preclinical to end-stage AD, and suggests that pathological changes in tau occur later in the disease process and are associated with clinical dementia (Thal et al., 1998).
CONCLUSION
AcknowledgementsöWe would like to acknowledge the assistance of the Sun Health Research Institute (Arizona, USA) in providing human brain material. This study was funded by the Department of Veterans A¡airs, The National Health and Medical Research Council and the Masonic Centenary Medical Research Foundation.
In this investigation, through the use of confocal imaging techniques, we de¢ned three major plaque types based on morphology. All plaque types were present in
REFERENCES
Armstrong, R.A., 1998. L-Amyloid plaques: Stages in life history or independent origin ? Dement. Geriatr. Cogn. Disord. 9, 227^238. Arnold, S.E., Hyman, B.T., Flory, J., Damasio, A.R., Van, H.G., 1991. The topographical and neuroanatomical distribution of neuro¢brillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer's disease. Cereb. Cortex 1, 103^116. Arriagada, P.V., Growdon, J.H., Hedly-Whyte, T., Hyman, B.T., 1992a. Neuro¢brillary tangles not senile plaques parallel duration and severity of Alzheimer's disease. Neurology 42, 631^639. Arriagada, P.V., Marzlo¡, K., Hyman, B.T., 1992b. Distribution of Alzheimer-type pathologic changes in nondemented elderly individuals matches the pattern in Alzheimer's disease. Neurology 42, 1681^1688. Benes, F.M., Farol, P.A., Majocha, R.E., Marotta, C.A., Bird, E.D., 1991. Evidence for axonal loss in regions occupied by senile plaques in Alzheimer cortex. Neuroscience 42, 651^660. Braak, H., Braak, E., 1991. Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol. (Berlin) 82, 239^259. Cruz, L., Urbanc, B., Buldyrev, S.V., Christie, R., Gomez-Isla, T., Havlin, S., McNamara, M., Stanley, H.E., Hyman, B.T., 1997. Aggregation and dissaggregation of senile plaques in Alzheimer's disease. Proc. Natl. Acad. Sci. USA 94, 7612^7616. Dickson, D.W., 1997. The pathogenesis of senile plaques. J. Neuropathol. Exp. Neurol. 56, 321^339. Dickson, D.W., Farlo, J., Davies, P., Crystal, H., Fuld, P., Yen, S.H., 1988. Alzheimer's disease : A double-labelling immunohistochemical study of senile plaques. Am. J. Pathol. 132, 86^101. Dickson, T.C., King, C.E., McCormack, G., Vickers, J.C., 1999. Neurochemical diversity of the dystrophic neurites in the early and late stages of Alzheimer's disease. Exp. Neurol. 156, 100^110. Hof, P.R., Morrison, J.H., 1994. The cellular basis of cortical disconnection in Alzheimer disease and related dementing conditions. In: Terry, R.D., Katzman, R., Bick, K.L. (Eds.), Alzheimer Disease. Raven, New York, pp. 197^229. Ikeda, S.I., Allsop, D., Glenner, G.G., 1989. Morphology and distribution of plaque and related deposits in the brains of Alzheimer's disease and control cases. An immunohistochemical study using amyloid L-protein antibody. Lab. Invest. 60, 113^122. Ikeda, S.I., Yanagisawa, M.B., Allsop, D., Glenner, G.G., 1990. Early senile plaques in Alzheimer's disease demonstrated by histochemistry, immunocytochemistry and electron microscopy. Hum. Pathol. 21, 1221^1226. Kato, S., Gondo, T., Hoshii, Y., Takahashi, M., Yamada, M., Ishihara, T., 1998. Confocal observation of senile plaques in Alzheimer's disease : Senile plaque morphology and relationship between senile plaques and astrocytes. Pathol. Int. 48, 332^340. Knowles, R.B., Wyart, C., Buldyrev, S.V., Cruz, L., Urbanc, B., Hasselmo, M.E., Stanley, H.E., Hyman, B.T., 1999. Plaque-induced neurite abnormalities : Implications for disruption of neural networks in Alzheimer's disease. Proc. Natl. Acad. Sci. USA 96, 5274^5279. Masliah, E., Mallory, M., Deernink, T., DeTeresa, R., Lamont, S., Miller, A., Terry, R.D., Carragher, R., Elisman, M., 1993a. Re-evaluation of the structural organization of neuritic plaques in Alzheimer's disease. J. Neuropathol. Exp. Neurol. 52, 619^632. Masliah, E., Mallory, M., Hansen, L., Alford, M., DeTeresa, R., Terry, R.D., 1993b. An antibody against phosphorylated neuro¢laments identi¢es a subset of damaged association axons in Alzheimer's disease. Am. J. Pathol. 142, 871^882. Mirra, S.S., Heyman, A., McKeel, D., Sumi, S.M., Crain, B.J., Brownlee, L.M., Vogel, F.S., Hughes, J.P., Van Belle, B., Berg, L., 1991. The consortium to establish a registry for Alzheimer's disease (CERAD). Part II. Standardization of the neuropathological assessment of Alzheimer's disease. Neurology 41, 479^486. Morris, J.C., McKeel, D.W.Jr., Storandt, M., Rubin, E.H., Price, J.L., Grant, E.A., Ball, M.J., Berg, L., 1991. Very mild Alzheimer's disease : informant-based clinical, psychometric, and pathologic distinction from normal aging. Neurology 41, 469^478. Morris, J.C., Storandt, M., McKeel, D.W., Rubin, E.H., Price, J.L., Grant, E.A., Berg, L., 1996. Cerebral amyloid deposition and di¡use plaques in `normal' aging: Evidence from presymptomatic and very mild Alzheimer's disease. Neurology 46, 707^719. Price, J.L., Morris, J.C., 1999. Tangles and plaques in nondemented aging and `preclinical' Alzheimer's disease. Ann. Neurol. 45, 358^368. Rozemuller, J.M., Eikelenboom, P., Stam, F.C., Beyreuther, K., Masters, C., 1989. A4 protein in Alzheimer's disease: Primary and secondary cellular events in extracellular amyloid deposition. J. Neuropathol. Exp. Neurol. 48, 674^691. Saunders, H.L., Dickson, T.C., Vickers, J.C., 1998. Immunolocalisation of L-amyloid precursor protein to plaque-associated synaptic alterations in the early and late stages of Alzheimer's disease. Alzheimer's Rep. 1, 111^119.
NSC 5014 16-7-01
Neuritic plaquesin Alzheimer's disease
107
Schmidt, M.L., Robinson, K.A., Lee, V.M-Y., Trojanowski, J.Q., 1995. Chemical and immunological heterogeneity of ¢brillar amyloid in plaques of Alzheimer's disease and Down's syndrome brains revealed by confocal microscopy. Am. J. Pathol. 147, 503^515. Selkoe, D.J., 1994. Alzheimer's disease: a central role for amyloid. J. Neuropathol. Exp. Neurol. 53, 438^447. Su, J.H., Cummings, B.J., Cotman, C.W., 1996. Plaque biogenesis in brain aging and Alzheimer's disease. I. Progressive changes in phosphorylation states of paired helical ¢laments and neuro¢laments. Brain Res. 739, 79^87. Terry, R.D., 1996. The pathogenesis of Alzheimer disease : An alternative to the amyloid hypothesis. J. Neuropathol. Exp. Neurol. 55, 1023^1025. Thal, D.R., Arendt, T., Waldmann, G., Holzer, M., Zedlick, D., Rub, U., Schober, R., 1998. Progression of neuro¢brillary changes and PHF-d in end-stage Alzheimer's disease is di¡erent from plaque and cortical microglial pathology. Neurobiol. Aging 19, 517^525. Ulrich, J., 1985. Alzheimer changes in nondemented patients younger than sixty-¢ve: Possible early stages of Alzheimer's disease and senile dementia of Alzheimer type. Ann. Neurol. 17, 273^277. Vickers, J.C., Chin, D., Edwards, A.-M., Sampson, V., Harper, C., Morrison, J., 1996. Dystrophic neurite formation associates with age-related L amyloid deposition in the neocortex : clues to the genesis of neuro¢brillary pathology. Exp. Neurol. 141, 1^11. Vickers, J.C., Dickson, T.C., Adlard, P.A., Saunders, H.L., King, C.E., McCormack, G., 2000. The cause of neuronal degeneration in Alzheimer's disease. Prog. Neurobiol. 60, 139^165. Vickers, J.C., Riederer, B.M., Marugg, R.A., Buee-Scherrer, V., Buee, L., Delacourte, A., Morrison, J.H., 1994. Alterations in neuro¢lament protein immunoreactivity in human hippocampal neurons related to normal aging and Alzheimer's disease. Neuroscience 62, 1^13. Wang, D., Munoz, D.G., 1995. Qualitative and quantitative di¡erences in senile plaque dystrophic neurites of Alzheimer's disease and normal aged brain. J. Neuropathol. Exp. Neurol. 54, 584^586. Wisniewski, H.M., Bancher, C., Barcikowska, M., Wen, G.Y., Currie, J., 1989. Spectrum of morphological appearance of amyloid deposits in Alzheimer's disease. Acta Neuropathol. 78, 337^347. Wisniewski, H.M., Terry, R.D., 1973. Reeaxamination of the pathogenesis of the senile plaque. Prog. Neuropathol. 2, 1^26. Yamaguchi, H., Hirai, S., Morimatsu, M., Shoji, M., Harigaya, Y., 1988. Di¡use type of senile plaques in the brains of Alzheimer-type dementia. Acta Neuropathol. 77, 113^119. Yasuhara, O., Kawamata, T., Aimi, Y., McGeer, E.G., McGeer, P.L., 1994. Two types of dystrophic neurites in senile plaques of Alzheimer disease and elderly non-demented cases. Neurosci. Lett. 171, 73^76. (Accepted 3 April 2001)
NSC 5014 16-7-01
Cyaan Magenta Geel Zwart