α-Secretase cleaved amyloid precursor protein (APP) accumulates in cholinergic dystrophic neurites...

a-Secretase cleaved amyloid precursor protein (APP)accumulates in cholinergic dystrophic neurites innormal, aged hippocampus

S.-Y. Yoon*, J.-U. Choi*, M.-H. Cho*, K.-M. Yang†, H. Ha†, I. J. Chung†, G. S. Cho* and D.-H. Kim*

*Department of Anatomy and Cell Biology, Cell Dysfunction Research Center (CDRC), Bio-Medical Institute ofTechnology (BMIT), University of Ulsan College of Medicine, and †National Forensic Service, Seoul, Korea

S.-Y. Yoon, J.-U. Choi, M.-H. Cho, K.-M. Yang, H. Ha, I. J. Chung, G. S. Cho and D.-H. Kim (2013) Neuropathologyand Applied Neurobiology 39, 800–816a-Secretase cleaved amyloid precursor protein (APP) accumulates in cholinergic dystrophic

neurites in normal, aged hippocampus

Aims: Dystrophic neurites are associated with b-amyloid(Ab) plaques in the brains of Alzheimer’s disease (AD)patients and are also found in some specific areas ofnormal, aged brains. This study assessed the molecularcharacteristics of dystrophic neurites in normal ageingand its difference from AD. Methods: We compared thedystrophic neurites in normal aged human brains (age20–70 years) and AD brains (Braak stage 4–6) by immu-nostaining against ChAT, synaptophysin, g-tubulin,cathepsin-D, Ab1–16, Ab17–24, amyloid precursorprotein (APP)-CT695 and APP-NT. We then tested thereproducibility in C57BL/6 mice neurone cultures.Results: In normal, aged mice and humans, we found anincrease in clustered dystrophic neurites of cholinergicneurones in CA1 regions of the hippocampus and layer IIand III regions of the entorhinal cortex, which are the

major and earliest affected areas in AD. These dystrophicneurites showed accumulation of sAPPa peptides cleavedfrom the amyloid precursor protein by a-secretase ratherthan Ab or C-terminal fragments. In contrast, Ab andAPP-CTFs accumulated in the dystrophic neurites in andaround Ab plaques of AD patients. Several experimentssuggested that the accumulation of sAPPa resulted fromageing-related proteasomal dysfunction. Conclusions:Ageing-associated impairment of the proteasomal systemand accumulation of sAPPa at cholinergic neurites in spe-cific areas of brain regions associated with memory couldbe associated with the normal decline of memory in agedindividuals. In addition, these age-related changes mightbe the most vulnerable targets of pathological insults thatresult in pathological accumulation of Ab and/or APP-CTFs and lead to neurodegenerative conditions such as AD.

Keywords: ageing, Alzheimer’s disease, APP, b-amyloid, dystrophic neurites, proteasome

Introduction

Neurofibrillary tangles (NFTs) and senile plaques are twopathological hallmarks of Alzheimer’s disease (AD) [1].

Extracellular plaques are primarily composed of b-amyloid peptides (Ab) derived from amyloid precursorprotein (APP). Dystrophic neurites are characterizedpathologically by aberrant neuritic sproutings or swollenneurites, and are widely associated with Ab plaques inAD [2]. Transgenic mice expressing mutant APP alsoexhibit dystrophic neurites surrounding Ab plaques[3,4], suggesting that amyloid deposition contributesto dystrophic neurites. However, diffuse forms of Ab

Correspondence: Dong-Hou Kim, Department of Anatomy and CellBiology, University of Ulsan College of Medicine, 88 Olympic-ro43-gil, SongPa-Gu, Seoul 138-736, Korea. Tel: +82 2 3010 4272;Fax: +82 2 3010 8063; E-mail: [email protected]

S.-Y.Y. and J.-U.C. equally contributes to this work.

800 © 2013 British Neuropathological Society

Neuropathology and Applied Neurobiology (2013), 39, 800–816 doi: 10.1111/nan.12032

mailto:[email protected]

plaques are not commonly associated with dystrophicneurites [5], indicating that Ab alone may not be suffi-cient to initiate the formation of dystrophic neurites.Interestingly, 22C11 (APP N-terminus antibody) andSMI-32 (nonphosphorylated neurofilament antibody)positive dystrophic neurites are also observed in normal,aged brains, especially in the hippocampus, which isimportant in memory formation and is a major affectedbrain region in AD [6,7]. This suggests that dystrophicneurites may precede Ab plaque formation. Thus,whether Ab deposition is a cause of dystrophic neuritesor whether dystrophic neurites affect Ab deposition is animportant unresolved issue.

Normal elderly people commonly experience ageing-dependent functional declines, including memoryimpairment [8,9]. The normal decline in cognition andmemory is related to the basal forebrain cholinergicsystem [10]. Degeneration and atrophy of cholinergicneurones are observed in normal, aged brains [11–13].One of the characteristic features of AD is also the loss ofcholinergic innervation [14,15]. Major cholinergic pro-jections to the cerebral cortex and hippocampus origi-nate from elements of the basal forebrain cholinergiccomplex, including the medial septum, ventral and hori-zontal diagonal band of Broca, and the basal nucleus ofMeynert [16].

Protein aggregation is a common, pathological eventthat is thought to be involved in the pathogenesis of neu-rodegenerative diseases such as AD, Parkinson’s disease,Huntington’s disease and amyotrophic lateral sclerosis[17]. Misfolded and aggregated proteins are generallydegraded by the ubiquitin–proteasome system (UPS) orthe autophagy–lysosome system. Interestingly, ubiquitindeposits are frequently observed in dystrophic neurites innormal, aged brains, suggesting impaired UPS function atthese sites [18,19].

Thus, dystrophic neurites, disturbances of the UPS andcholinergic deficits are characteristic features of normal,aged brains that have been implicated in the normaldecline in memory and learning, and may further contrib-ute to the progression to AD. In this context, it isimportant to distinguish normal ageing from AD inunderstanding the precise features of disease and design-ing therapeutic strategies. However, few studies haveaddressed the reciprocal relationship between dystrophicneurites and the cholinergic system in normal, aged brainregions, including the hippocampus. Hence, we herestudied it and found that a-secretase-cleaved APP accu-

mulates in cholinergic dystrophic neurites in normal,aged hippocampus via age-related decline of proteasomesystem.

Materials and methods

Human tissues

Normal, aged human tissue, consisting of brains from 16women and 32 men (age below 20–30 years; 19 cases,age 40–50 years; 16 cases, age 50–70 years; 13 cases)who had no clinical status of dementia, epilepsy or braintrauma, was provided by the Korean National ForensicService and the brains were proved to have no pathologi-cal findings of AD such as NFTs and Ab plaques. Accord-ing to AD diagnostic criteria (Diagnostic and StatisticalManual of Mental Disorders, Fourth Edition), short-termor long-term memory loss is essential for diagnosis forAD [20]. People who donated normal aged brain tissueswere reported that there was no reported long-termor short-term memory decline. Brain tissue from ADpatients (n = 4 brains; patient ages, 57–82 years; Braakstage, 4–6) was provided by the Netherlands Brain Bank.Paraffin-embedded sections from the hippocampus andentorhinal cortex were prepared for use in proceduresdescribed below. As the brains of AD and normal caseswere derived from different sources, there was race differ-ence that is AD from Caucasian and normal cases fromAsian. Although there was race difference, other diag-nostic criteria or tissue preservation was controlled in asimilar condition. All procedures were approved by theInstitutional Review Board of the Asan Medical Center inSeoul, Korea.

Animal tissues

Changes in the hippocampus and entorhinal cortex asso-ciated with normal ageing were evaluated in brain tissuesfrom 4- or 6-month-old (adult), 12-month-old (middleage) and 15-month-old (aged) C57BL/6 mice. Animalswere maintained at a constant ambient temperature(22°C � 1°C) with a 12 h:12 h light:dark cycle and freeaccess to water and food. Mice were sacrificed after anaes-thetization with an intraperitoneal injection of pentobar-bital sodium (40 mg/kg). Mice were decapitated using aguillotine and brains were immediately dissected onice and subsequently embedded in paraffin. Dewaxedparaffin-embedded sections (5 mm thick) were mounted

APP-positive cholinergic dystrophic neurites in aged hippocampus 801

NAN 2013; 39: 800–816© 2013 British Neuropathological Society

on glass slides and processed for immunohistochemistry.All described procedures were approved by the Institu-tional Animal Care and Use Committee of the AsanInstitute for Life Sciences in Seoul, Korea.

Immunohistochemistry

Paraffin-embedded sections collected throughout thehippocampal formation were incubated overnight at4°C with the following primary antibodies [diluted inphosphate-buffered saline (PBS) containing 2% normalgoat serum and 0.3% Triton X-100]: mouse anti-APP-N-terminus (NT, 1:100; Millipore, Bedford, MA, USA),SMI-32 (1:1000; Sternberger monoclonals), rabbitanti-ChAT (1:100; Millipore), mouse anti-synaptophysin(1:500; Sigma, St. Louis, MO, USA), mouse anti-g-tubulin(1:100; Sigma), rabbit anti-cathepsin-D (1:100; DAKO,Carpinteria, CA, USA), mouse anti-Ab1–16 (clone6E10, 1:500; Signet, Emeryville, CA, USA), mouse anti-Ab17–24 (clone 4G8, 1:100; Signet), rabbit anti-APP-CT695 (1:100; Invitrogen, Carlsbad, CA, USA), mouseanti-APP-NT (clone 22C11, 1:200; Chemicon, Temecula,CA, USA) and rat anti-sAPPa (4B4, kindly provided byProfessor Stefan F Lichtenthaler [21]). Sections were thenwashed three times with PBS. For immunofluorescence,sections were incubated for 30 min at room temperaturein the appropriate fluorescein isothiocyanate (FITC)- orCy3-conjugated secondary antibodies (Jackson Immu-noResearch Laboratories, West Grove, PA, USA) washedagain, placed on gelatinized glass slides, air dried andcoverslip-mounted with FluorSave mounting medium(Calbiochem, San Diego, CA, USA). Sections were alsostained with thioflavin-S to reveal AD plaques.

Primary cortical neurone culture andlactacystin treatment

Primary mouse cortical neurone cultures has beendescribed previously [22]. Cells, harvested from cortices of16-day-old mouse embryos, were plated in neurobasalmedium containing B27 supplement (Gibco BRL,Carlsbad, CA, USA), 25 mM glutamate, 500 mMl-glutamine and penicillin (100 IU/ml)/streptomycin(50 mg/ml). The cells were then maintained at 37°C in a5% CO2 incubator during designated times. Two days afterplating, half the medium was removed and replaced withfresh medium without glutamate, the medium waschanged every 4 days. At 10 days after plating, proteasome

inhibitor lactacystin (10 mM) was treated in culturesfor 24 h.

Statistical analysis

Data are presented as means � SEMs and were analysedusing Student’s t-test. A P-value < 0.05 was considered tobe statistically significant.

Results

Accumulation of APP-positive cholinergicdystrophic neurites in the hippocampus andentorhinal cortex of aged mice

As ageing progressed, dystrophic neurites that stainedwith an antibody recognizing the N-terminus of APP(APP-NT) appeared in the hippocampus and entorhinalcortex (Figure 1, arrows). APP-NT (+) dystrophic neuriteswere not detectable in brain sections from mice youngerthan 6 months of age (Figure 1). APP-NT (+) dystrophicneurites frequently appeared after 12 months and furtherincreased at 15 months. APP (+) dystrophic neuritesappeared as clusters, suggesting that they might bebundles of the same group of neurones. These clusterswere mainly localized in the CA1 area of the hippocampus,where they were distributed in the stratum oriens, stratumradiatum, and lacunosum molecular layers (Figure 1,arrows). They were also distributed in layer II/III of theentorhinal cortex (Figure 1, arrows). To investigate theseAPP (+) cluster is dystrophic neurites found in aged humanbrain as described previously [6,7], we stained the15-month-old brain sections with SMI-31, -32, -34, whichdetect phosphorylated or nonphosphorylated neurofila-ments, and PHF-1. As a result, the distended clustersstained with SMI-32 (nonphosphorylated neurofilaments,Figure 1C) but not with SMI-31, -34 and PHF-1, indicatingthat these clustered APP and nonphosphorylated neuro-filament (+) dystrophic neurites were similar to dystrophicneuritis of aged human brain.

Various studies have linked age-dependent cognitivedecline to deficits in cholinergic synaptic transmission andplasticity [23–25]. Accordingly, we tested whether APP(+), clustered dystrophic neurites were cholinergic inorigin. We found that these neurites in the hippocampusand entorhinal cortex were colocalized with choline acetyl-transferase (ChAT) immunoreactivity (Figure 2, arrows),indicating that they originated primarily from cholinergic

802 S.-Y. Yoon et al.


Figure 1. Age-dependent amyloid precursor protein (APP) (+) dystrophic neurites in the hippocampus and entorhinal cortex. (A) Coronalsections of the hippocampus and entorhinal cortex from 6-, 12- and 15-month-old mice were immunostained with antibody 22C11, whichrecognizes the N-terminus of APP. After 12 months of age, APP (+), clustered dystrophic neurites (arrows) were apparent in the stratumradiatum, lacunosum molecular and stratum oriens layers of the CA1 region of the hippocampus and in layer II/III of the entorhinal cortex.(B) The number of APP (+) dystrophic clusters in the CA1 area of hippocampus (6 months n = 4, 12 months n = 5, 15 months n = 5,*P-value < 0.05). (C) SMI-32 labelling in hippocampus CA1 of the 15-month-old mouse brain. Arrows indicate bouton en passant likeprocess (arrows). Scale bar = 50 mm.



neurones. These cholinergic dystrophic neurites were alsoimmunoreactive for synaptophysin, a marker of synapticvesicles (Figure 3, arrows), but did not stain with anantibody recognizing MAP2 (microtubule-associatedprotein 2), which selectively labels dendritic trees (data notshown), indicating that they were dystrophic axon termi-nals or swollen axons with accumulated synaptic vesicles.

Interestingly, these age-related dystrophic neuriteswere also positive for g-tubulin and cathepsin-D (Figure 4,arrows). g-Tubulin, an atypical member of the tubulinsuperfamily, has been identified in many types of eukaryo-tic cells and is believed to play a key role in the nucleationof microtubules [26]. In addition, g-tubulin is a marker ofaggresomes, which are structures composed of aggre-

gated proteins that form under conditions of increasedaccumulation of aberrant proteins [27]. Cathepsin-D, alysosomal hydrolase and a marker of lysosomes, was alsodetected under conditions in which aberrant proteinsaccumulated. These features suggest that APP accumula-tion in age-related clusters of cholinergic dystrophic neu-rites in the hippocampus and entorhinal cortex may berelated to proteasomal dysfunction.

Identification of accumulated APPs asa-secretase-cleaved forms of APP

Amyloid precursor protein can be cleaved by a-, b- org-secretases, generating a variety of APP fragments. As

Figure 2. Cholinergic dystrophic neurites in the aged hippocampus and entorhinal cortex. Coronal sections of the hippocampus andentorhinal cortex from 15-month-old mice were double-immunolabelled with an anti-ChAT antibody, and with 22C11, to detect amyloidprecursor protein (APP)-NT, or with 6E10, to detect Ab1–16 of APP. 22C11 (+) and 6E10 (+) dystrophic neurites colocalized with ChATimmunostaining (arrows), indicating their cholinergic nerve origin. Scale bar = 50 mm.



Figure 3. Synaptic vesicle accumulation in cholinergic dystrophic axons of the aged hippocampus. Coronal sections of hippocampus from4- and 15-month-old mice were double-labelled with antibodies against ChAT and synaptophysin, a synaptic vesicle glycoprotein thatnormally localizes to axon terminals. There were no dystrophic neurites in 4-month-old mouse brains; however, ChAT (+) dystrophicneurites were colocalized with synaptophysin (arrows) in 15-month-old mouse brains, indicating that cholinergic dystrophic neurites areaxons and accumulate with synaptic vesicles. (A) Lower magnification; (B) higher magnification. Scale bar = 50 mm.



Figure 4. Accumulation of g-tubulin and cathepsin-D in dystrophic neurites of the aged hippocampus. Coronal hippocampal sections from6-, 12- and 15-month-old mice were double-labelled with antibodies against ChAT and g-tubulin, an aggresome marker (arrowheads) (A).Scale bar = 20 mm. The right-most panel shows a higher magnification image of (A). Amyloid precursor protein (APP)-NT immunoreactive,clustered dystrophic neurites were colocalized (arrows) with cathepsin-D, a lysosome marker (B), indicating that dystrophic neuritesaccumulated aggresomes and lysosomes. Scale bar = 50 mm.



shown in Figures 1, 2 and 4, the observed APP stainedwith an antibody raised against an N-terminal epitope ofAPP (22C11) corresponding to amino acids 60–100,and with an antibody (6E10) raised against N-terminalamino acids 1–16 of the Ab peptide. This suggests thatthese APP signals could correspond to full-length APP ora-/b-cleaved APP fragments. Because Ab deposition is aprominent feature of AD and can also be observed innormal, aged individuals, we suspected these APP dys-trophic neurites might contribute to Ab deposition. Tomore precisely define the identity of the APP in the dys-trophic neurites, we further immunostained sectionswith antibodies targeting other APP epitopes and sec-tions were immunostained with alternative DAB methodand higher-power images of dystrophic neurites usingconfocal microscope were obtained. Interestingly,although the dystrophic neurites stained with the anti-body directed against Ab1–16 (6E10), they did not reactwith an antibody (4G8) against amino acids 17–24 ofAb or with an antibody (CT695) that recognizes theC-terminus of APP (Figure 5 and Figures S1 and S2).Recently, Amadoro et al. showed that the 6E10 antibodycan recognize murine APP and sAPPa [28]. This indi-cates that accumulated APP in normal, aged mice wasnot full-length APP, Ab or CTFs, but was instead ana-secretase-cleaved form of APP (sAPPa).

Similar to the features observed in aged mice, APP (+)dystrophic neurites were observed in the human hippoc-ampus during the normal ageing process (Figure 6). Theseneurites were detected at an early age (even in the 20s)and increased with age, especially in the stratum molecu-lar layer of the dentate gyrus. These APP (+) dystrophicneurites, like those in mice, stained with 22C11, 6E10 andsAPPa, but not with 4G8 or CT695 (Figure 7), indicatingthe accumulation of sAPPa. They also stained forg-tubulin and cathepsin-D, suggesting an ageing-relatedprotein aggregation process in these dystrophic neurites,possibly reflecting a decline in proteasomal function.

Accumulation of different species of APP in theAD hippocampus

In contrast to the normal, aged human hippocampus,where sAPPa accumulated in dystrophic neurites, Abplaques in the hippocampus of AD patients contained dys-trophic neurites that were immunopositive for all APPantibodies. Dystrophic neurites in and around Ab plaquesalso contained cathepsin-D and g-tubulin (Figure 8). This

indicates that full-length APP, APP-NTs, Ab peptides andAPP-CTFs could all accumulate in dystrophic neurites inand around Ab plaques, suggesting that aberrant process-ing of APP may contribute to Ab deposition.

Proteasome inhibition mimics ageing-relatedchanges in neuronal culture

It has been suggested that impaired proteasomal systemfunction is associated with normal brain ageing [29,30].To address the role of proteasome function in ageing-related neuronal changes, we treated neurones withlactacystin, a proteasome inhibitor. Treatment with lacta-cystin induced an increase in cathepsin-D (+) andg-tubulin (+) dystrophic neurites (Figure S3), indicatingthat lysosomes and aggresomes were accumulated inthese dystrophic neurite foci, similar to what was observedin the aged hippocampus of mice and humans (Figures 4and 6 and Figure S1).

We also tested whether sAPPa accumulated in dys-trophic neurites after lactacystin treatment as it did in thenormal, aged hippocampus (Figures 5 and 7). Interest-ingly, dystrophic neurites with 6E10 signals increased afterlactacystin treatment and colocalized with cathepsin-D(Figures S3 and S4 arrowheads). However, only a few dys-trophic neurites with 4G8 signals were evident and theydid not colocalize with cathepsin-D (Figure S4). Dystrophicneurites showed substantial immunostaining for APP-NT,but APP-CT695 signals were less intense and did not colo-calize with APP-NT (Figure S4). These results show thatsAPPa rather than Ab or CTFs accumulated in dystrophicneurites after lactacystin treatment, similar to what wasobserved in the normal, aged hippocampus (Figures 5and 7).

The fact that proteasome inhibition mimicked ageing-related changes, but not AD-like changes, in mice andhumans is interesting because it may imply that UPSimpairment plays an important role in the ageing processin the brain. However, inhibition of the proteasomalsystem alone may not be sufficient to induce formation ofthe Ab- and CTF-accumulating dystrophic neurites char-acteristic of AD pathology. Thus, some additional patho-logical factors are likely involved in the formation ofAD-specific dystrophic neurites.

Discussion

We here showed that APP (+) and ChAT (+), clustereddystrophic neurites increased with ageing in the



Figure 5. Identification of amyloid precursor protein (APP) species as a-secretase-cleaved forms that are accumulated in dystrophicneurites of aged mice. (A) Cathepsin-D (+) dystrophic neurites were immunopositive for the 6E10 antibody, which detects Ab1–16. However,neither cathepsin-D (+) nor g-tubulin (+) dystrophic neurites were immunopositive for the 4G8 antibody, which detects Ab17–24, or theAPP-CT695 antibody, which detects the C-terminus of APP. Thus, dystrophic neurites of normal, aged hippocampus accumulatea-secretase-cleaved APP peptides, not full-length APP, b-amyloid (Ab) or CTFs. Scale bar = 50 mm. (B) Schematic diagram of APP speciesand anti-APP antibodies, indicating the distinct APP epitopes detected by each antibody.



Figure 6. Increases in dystrophic neurites with normal ageing in the human hippocampus. (A) Coronal sections of the normal humanhippocampus were immunostained for amyloid precursor protein (APP)-NT. In the lacunosum molecular layer of the dentate gyrus,APP-NT (+) dystrophic neurites (arrows) appeared in individuals as young as 23 years and increased as ageing progressed. (B) Dystrophicneurites of the normal human hippocampus also stained for g-tubulin (arrows), and g-tubulin staining increased with age. Dystrophicneurites in and around the thioflavin-S (+) senile plaques of the Alzheimer’s disease (AD) hippocampus were immunopositive for g-tubulin(arrows). Scale bar = 50 mm. (C) The number of APP (+) dystrophic neurites in the normal aged hippocampus (age 20–30 years; 19 cases,age 40–50 years; 16 cases, age 50–70 years; 13 cases, *P-value < 0.05).



Figure 7. Accumulation of a-secretase-cleaved forms of APP in dystrophic neurites of the normal, aged human hippocampus. Dystrophicneurites in coronal hippocampal sections from a 56-year-old male, like those of aged mice, also immunostained for cathepsin-D andg-tubulin, indicating that aggresomes and lysosomes are accumulated in these dystrophic neurites. These dystrophic neurites (arrows) wereimmunostained by 6E10, sAPPa, but not by 4G8 or APP-CT695, indicating that the accumulated APPs are a-secretase-cleaved forms. Scalebar = 50 mm.



hippocampus and entorhinal cortex of mouse and humanbrains. These APP (+) cholinergic dystrophic neuritesaccumulated sAPPa fragments and were mostly distrib-uted in stratum radiatum and lacunosum molecularlayers of the hippocampus CA1 area and the stratummolecular layer of the dentate gyrus, which is the regionmost vulnerable to degeneration in AD [31]. These resultsare similar in appearance and distribution to previousreports that have shown the age-related occurrence ofclusters of granular deposits predominantly in the hippoc-ampus of C57BL/6 mice and senescence-acceleratedmouse (SAM) P8 strain and Wistar rat [32,33]. Interest-ingly, Ab (+) and CTF (+) dystrophic neurites appearedonly in AD brains, not in normal, aged brains (Figure 8).Several studies have suggested that alterations in synaptic

transmission, especially in the cholinergic network [34],may contribute to age-dependent neuronal dysfunction[23–25,35–37]. Several patterns of axonal abnormalities,which is enlarged, frequently argentophilic nodular vari-cosities and clusters of neurites sometimes associated withvariable amounts of amyloid, were observed in agedbrains [38]. Although no significant decline in neuronalnumber in the nucleus basalis Meynert complex of thebasal forebrain is found from the normal aged brains [39],a strong correlation with the degeneration of cholinergicafferents and neuritic plaque formation was revealed [40].Thus, it is highly likely that the observed neuritic changesmay contribute significantly to the synaptic and behav-ioural impairments frequently observed in the aged popu-lation, and may further lead to amyloid deposition during

Figure 8. Accumulation of b-amyloid (Ab), CTFs and full-length amyloid precursor protein (APP) in the senile plaques in the hippocampusof Alzheimer’s disease (AD) patients. Dystrophic neurites in and around thioflavin-S (+) senile plaques in coronal hippocampal sections ofBraak stage V/VI were immunostained by APP-NT, 4G8 and APP-CT695 antibodies. This indicates that Ab, CTFs and full-length APP areaccumulated in the dystrophic neurites in and around senile plaques of the AD hippocampus. Scale bar = 50 mm.



ageing [41]. It is thought that axonopathy or neuronalinjury often causes neuritic dystrophy. Factors such asimpaired axonal transport [42], increased membrane fra-gility [43] or cytoskeletal alterations [44] that occur inneurodegenerative diseases have been suggested to facili-tate the formation of dystrophic neurites.

Antibodies that recognize APP are often used to detectdystrophic neurites surrounding amyloid plaques in ADbrains [45]. Moreover, the presence of dystrophic neuritessurrounding amyloid plaques in mutant APP-expressingmouse models implies that the formation of dystrophicneurites is secondary to Ab deposition [46]. On the otherhand, recent studies have also demonstrated that dys-trophic neurites are generated during the normal ageingprocess before detectable axonal or neuronal injuries orother neurodegenerative features, and in the absence ofsenile plaques [47–49]. These observations, which aresimilar to those reported here, suggest that dystrophic neu-rites in normal ageing are commonly linked to a decline inlearning and memory [8,9]. However, it has not beenclearly established whether these dystrophic neurites are aconsequence or a cause of amyloid deposition. Moreover,previous studies have not clearly addressed which speciesof APPs accumulate in dystrophic neurites in normal, agedpopulations and in AD patients. Thus, the precise molecu-lar composition of APPs in dystrophic neurites, as well astemporal and spatial aspects of their formation are allimportant factors in determining the pathophysiologicalsignificance of the occurrence of dystrophic neurites.

Our present study showed that dystrophic neuritesstained positively with an antibodies that detects theN-terminal Ab1–16 epitope of APP or sAPPa specifically[21], but were not stained by APP C-terminus- or Ab17–24-recognizing antibodies in normal aged brains, leadingus to conclude that sAPPa species are accumulated innormal, aged hippocampus and entorhinal cortex, twokey areas in memory formation and maintenance. sAPPahas been long thought to have a positive role in synapticplasticity, memory, neuroprotection and neurotrophicfunctions [50]; hence, sAPPa accumulation in dystrophicneurites reflects its impaired function and loss of its ben-eficial effects, leading to the normal ageing-related declinein memory function. However, dystrophic neurites in andaround Ab plaques of AD patients were also positive forthe APP C-terminus and Ab17–24, indicating that Aband CTFs were also accumulated in dystrophic neurites ofAD patients. Thus, it is not likely that extracellular Abdeposits are simply derived from the APP found within

dystrophic neurites in normal, aged brains. This suggeststhat additional pathological mechanisms may lead to thesubsequent generation of Ab plaques in dystrophic neur-ites. An important and interesting issue yet to be resolvedis the identification of factors that contribute to the pro-gression of dystrophic neurites to Ab plaques, and thuslead to AD instead of normal ageing.

Although we think that sAPPa is the main speciesaccumulated in the dystrophic neurites of normal agedbrains, there is a possibility that sAPPb′ may be accumu-lated as 6E10 antibody reacts to an epitope between 4 and9 of Ab. BACE1 cleaves APP at b′-site, glutamate 11 ofAb, in addition to b-site, aspartate 1 of Ab [51,52]. Hence,APP species detected by 22C11 and 6E10 antibodies couldbe sAPPa or sAPPb′. However, if b′-cleavage preferentiallyoccurs in normal aged brains, b′-amyloid may be accumu-lated and plaques may be present but 4G8 antibody orthioflavin-S could not detect any plaques. Hence, we thinkthat a-cleavage rather than b′-cleavage predominates innormal aged brains and that some factors provoking b- orb′-cleavage may lead to diseased status.

In this study, we found that APP (+) dystrophic neuritesin the aged hippocampus were immunoreactive forg-tubulin and cathepsin-D (Figure 4). g-Tubulin is amarker of aggresomes, which commonly form as a cellu-lar response to an excess of misfolded or aberrant proteinsunder conditions in which the degradation capacity of theUPS is exceeded or impaired [53]. The aggregation of pro-teins and the formation of inclusion bodies in neurode-generative diseases are regarded as aggresome-relatedprocesses [54,55]. The formation of multiple aggregatesand redistribution of g-tubulin were reported in dys-trophic neurites and the extracellular space in rotenone-treated neurones [56]. Aggregation of g-tubulin was alsoreported in lactacystin-treated dopaminergic neurones[55], suggesting that the formation of aggresomes may bea cellular strategy for isolating or degrading aberrant pro-teins when the UPS is impaired [53]. Thus, g-tubulin depo-sition in cholinergic nerve terminals in the hippocampusand entorhinal cortex with ageing reflects gradual UPSimpairment, which may then proceed to neurodegenera-tive conditions such as AD. Hence, the ageing-associatedaccumulation of APP, g-tubulin and cathepsin-D may allbe related to impairment of the UPS. The UPS is importantin neurodevelopment, synaptic function, and the plastic-ity and survival of neurones [57]. Neuropathologicalinclusions in various neurodegenerative diseases, includ-ing Ab and NFTs in AD, contain ubiquitinated proteins,



suggesting impairment of the UPS [58,59]. UPS impair-ment precedes the onset of neurodegenerative diseasessuch as AD and PD [60,61], suggesting a causative roleearly in the pathogenesis. Moreover, the UPS becomesimpaired in the normal brain, including in the hippocam-pus, with ageing [29,30]. Proteasome inhibition arrestsneurite outgrowth and causes neurodegeneration [62].

Our study revealed that sAPPa accumulates in thecholinergic dystrophic neurites of the hippocampus andentorhinal cortex of normal, aged humans, whereas otherAbs or CTFs accumulate in the dystrophic neurites ofsenile plaques of AD patients. This distinction couldrepresent a molecular diagnostic tool for differentiatingbetween normal ageing and AD. This ageing-related APPaccumulation may involve UPS impairment; however,UPS impairment alone is not sufficient to lead to AD. Thus,attenuating UPS impairment may help prevent normalageing-related memory decline and eliminate other, as yetunknown, causative factors that lead to AD conditions,thereby helping to prevent or alleviate AD.

Acknowledgements

This study was supported by grants from the NationalResearch Foundation of Korea funded by the Korean gov-ernment (2008-0062466); the Basic Science ResearchProgram through the National Research Foundation ofKorea (NRF) funded by the Ministry of Education,Science and Technology (2010-0002859); the AsanInstitute for Life Sciences, Seoul, Korea (10-067). Wethank the National Forensic Service (NFS) and the Neth-erlands Brain Bank (NBB) for supplying human brainmaterials and also thank the donors and their relativesfor the kind gift of human brains for neuropathologicalstudies.

Author contributions

S.-Y.Y. and D.-H.K. designed the study. S.-Y.Y., J.-U.C.,M.-H.C. and G.S.C. carried out the experimental work andcarried out the histopathological analysis of the samples.K.-M.Y. and H.H. carried out the clinical evaluation of thesamples. K.-M.Y., H.H. and I.J.C. provided the tissues.S.-Y.Y., J.-U.C. and D.-H.K. wrote the paper.

References

1 Gomez-Isla T, Hollister R, West H, Mui S, Growdon JH,Petersen RC, Parisi JE, Hyman BT. Neuronal loss corre-

lates with but exceeds neurofibrillary tangles in Alzheim-er’s disease. Ann Neurol 1997; 41: 17–24

2 Dickson DW, Farlo J, Davies P, Crystal H, Fuld P, Yen SH.Alzheimer’s disease. A double-labeling immunohisto-chemical study of senile plaques. Am J Pathol 1988; 132:86–101

3 Games D, Adams D, Alessandrini R, Barbour R, Ber-thelette P, Blackwell C, Carr T, Clemens J, Donaldson T,Gillespie F. Alzheimer-type neuropathology in transgenicmice overexpressing V717F beta-amyloid precursorprotein. Nature 1995; 373: 523–7

4 Holcomb L, Gordon MN, McGowan E, Yu X, Benkovic S,Jantzen P, Wright K, Saad I, Mueller R, Morgan D,Sanders S, Zehr C, O’Campo K, Hardy J, Prada CM,Eckman C, Younkin S, Hsiao K, Duff K. AcceleratedAlzheimer-type phenotype in transgenic mice carryingboth mutant amyloid precursor protein and presenilin 1transgenes. Nat Med 1998; 4: 97–100

5 Joachim CL, Morris JH, Selkoe DJ. Diffuse senile plaquesoccur commonly in the cerebellum in Alzheimer’sdisease. Am J Pathol 1989; 135: 309–19

6 Vickers JC, Chin D, Edwards AM, Sampson V, Harper C,Morrison J. Dystrophic neurite formation associated withage-related beta amyloid deposition in the neocortex:clues to the genesis of neurofibrillary pathology. ExpNeurol 1996; 141: 1–11

7 Su JH, Cummings BJ, Cotman CW. Plaque biogenesisin brain aging and Alzheimer’s disease. II. Progressivetransformation and developmental sequence of dys-trophic neurites. Acta Neuropathol (Berl) 1998; 96:463–71

8 Winocur G, Moscovitch M. A comparison of cognitivefunction in community-dwelling and institutionalized oldpeople of normal intelligence. Can J Psychol 1990; 44:435–44

9 Gallagher M, Rapp PR. The use of animal models to studythe effects of aging on cognition. Annu Rev Psychol 1997;48: 339–70

10 Bartus RT, Dean RL, Beer B, Lippa AS. The cholinergichypothesis of geriatric memory dysfunction. Science1982; 217: 408–14

11 Fischer W, Chen KS, Gage FH, Bjorklund A. Progressivedecline in spatial learning and integrity of forebraincholinergic neurons in rats during aging. Neurobiol Aging1992; 13: 9–23

12 Madhusudan A, Sidler C, Knuesel I. Accumulation ofreelin-positive plaques is accompanied by a decline inbasal forebrain projection neurons during normal aging.Eur J Neurosci 2009; 30: 1064–76

13 Stroessner-Johnson HM, Rapp PR, Amaral DG. Choliner-gic cell loss and hypertrophy in the medial septal nucleusof the behaviorally characterized aged rhesus monkey.J Neurosci 1992; 12: 1936–44

14 Coyle JT, Price DL, DeLong MR. Alzheimer’s disease: adisorder of cortical cholinergic innervation. Science1983; 219: 1184–90



15 Davies P, Maloney AJ. Selective loss of central cholinergicneurons in Alzheimer’s disease. Lancet 1976; 2: 1403

16 Mufson EJ, Ginsberg SD, Ikonomovic MD, DeKosky ST.Human cholinergic basal forebrain: chemoanatomy andneurologic dysfunction. J Chem Neuroanat 2003; 26:233–42

17 Ross CA, Poirier MA. Protein aggregation and neurode-generative disease. Nat Med 2004; 10 (Suppl.): S10–17

18 Dickson DW, Crystal HA, Mattiace LA, Masur DM, BlauAD, Davies P, Yen SH, Aronson MK. Identification ofnormal and pathological aging in prospectively studiednondemented elderly humans. Neurobiol Aging 1992; 13:179–89

19 Dickson DW, Wertkin A, Kress Y, Ksiezak-Reding H, YenSH. Ubiquitin immunoreactive structures in normalhuman brains. Distribution and developmental aspects.Lab Invest 1990; 63: 87–99

20 Cummings JL. Alzheimer’s disease. N Engl J Med 2004;351: 56–67

21 Kuhn PH, Wang H, Dislich B, Colombo A, Zeitschel U,Ellwart JW, Kremmer E, Rossner S, Lichtenthaler SF.ADAM10 is the physiologically relevant, constitutivealpha-secretase of the amyloid precursor protein inprimary neurons. EMBO J 2010; 29: 3020–32

22 Ahn J, Choi J, Joo CH, Seo I, Kim D, Yoon SY, Kim YK, LeeH. Susceptibility of mouse primary cortical neuronal cellsto coxsackievirus B. J Gen Virol 2004; 85: 1555–64

23 deToledo-Morrell L, Geinisman Y, Morrell F. Age-dependent alterations in hippocampal synaptic plasticity:relation to memory disorders. Neurobiol Aging 1988; 9:581–90

24 Campbell LW, Hao SY, Thibault O, Blalock EM, LandfieldPW. Aging changes in voltage-gated calcium currents inhippocampal CA1 neurons. J Neurosci 1996; 16:6286–95

25 Smith TD, Adams MM, Gallagher M, Morrison JH,Rapp PR. Circuit-specific alterations in hippocampalsynaptophysin immunoreactivity predict spatial learningimpairment in aged rats. J Neurosci 2000; 20: 6587–93

26 Oakley BR. gamma-Tubulin. Curr Top Dev Biol 2000; 49:27–54

27 Zhao LY, Liao D. Sequestration of p53 in the cytoplasm byadenovirus type 12 E1B 55-kilodalton oncoprotein isrequired for inhibition of p53-mediated apoptosis. J Virol2003; 77: 13171–81

28 Amadoro G, Corsetti V, Ciotti MT, Florenzano F, CapsoniS, Amato G, Calissano P. Endogenous Abeta causes celldeath via early tau hyperphosphorylation. NeurobiolAging 2011; 32: 969–90

29 Keller JN, Gee J, Ding Q. The proteasome in brain aging.Ageing Res Rev 2002; 1: 279–93

30 Paz Gavilan M, Vela J, Castano A, Ramos B, del Rio JC,Vitorica J, Ruano D. Cellular environment facilitatesprotein accumulation in aged rat hippocampus. NeurobiolAging 2006; 27: 973–82

31 West MJ, Coleman PD, Flood DG, Troncoso JC. Differencesin the pattern of hippocampal neuronal loss in normalageing and Alzheimer’s disease. Lancet 1994; 344:769–72

32 Kuo H, Ingram DK, Walker LC, Tian M, Hengemihle JM,Jucker M. Similarities in the age-related hippocampaldeposition of periodic acid-schiff-positive granules in thesenescence-accelerated mouse P8 and C57BL/6 mousestrains. Neuroscience 1996; 74: 733–40

33 Knuesel I, Nyffeler M, Mormede C, Muhia M, Meyer U,Pietropaolo S, Yee BK, Pryce CR, LaFerla FM, MarighettoA, Feldon J. Age-related accumulation of Reelin inamyloid-like deposits. Neurobiol Aging 2009; 30: 697–716

34 Winkler J, Suhr ST, Gage FH, Thal LJ, Fisher LJ. Essentialrole of neocortical acetylcholine in spatial memory.Nature 1995; 375: 484–7

35 Nicholson DA, Yoshida R, Berry RW, Gallagher M, Gein-isman Y. Reduction in size of perforated postsynapticdensities in hippocampal axospinous synapses and age-related spatial learning impairments. J Neurosci 2004;24: 7648–53

36 Morrison JH, Hof PR. Life and death of neurons in theaging cerebral cortex. Int Rev Neurobiol 2007; 81: 41–57

37 Andrews-Hanna JR, Snyder AZ, Vincent JL, Lustig C,Head D, Raichle ME, Buckner RL. Disruption of large-scale brain systems in advanced aging. Neuron 2007; 56:924–35

38 Price DL, Cork LC, Struble RG, Whitehouse PJ, Kitt CA,Walker LC. The functional organization of the basal fore-brain cholinergic system in primates and the role of thissystem in Alzheimer’s disease. Ann N Y Acad Sci 1985;444: 287–95

39 Bigl V, Arendt T, Fischer S, Werner M, Arendt A. Thecholinergic system in aging. Gerontology 1987; 33:172–80

40 Arendt T, Bigl V, Tennstedt A, Arendt A. Neuronal loss indifferent parts of the nucleus basalis is related to neuriticplaque formation in cortical target areas in Alzheimer’sdisease. Neuroscience 1985; 14: 1–14

41 Sperling RA, Laviolette PS, O’Keefe K, O’Brien J, RentzDM, Pihlajamaki M, Marshall G, Hyman BT, Selkoe DJ,Hedden T, Buckner RL, Becker JA, Johnson KA. Amyloiddeposition is associated with impaired default networkfunction in older persons without dementia. Neuron2009; 63: 178–88

42 Stokin GB, Lillo C, Falzone TL, Brusch RG, Rockenstein E,Mount SL, Raman R, Davies P, Masliah E, Williams DS,Goldstein LS. Axonopathy and transport deficits early inthe pathogenesis of Alzheimer’s disease. Science 2005;307: 1282–8

43 Praprotnik D, Smith MA, Richey PL, Vinters HV, Perry G.Plasma membrane fragility in dystrophic neurites insenile plaques of Alzheimer’s disease: an index of oxida-tive stress. Acta Neuropathol (Berl) 1996; 91: 1–5



44 Dickson TC, King CE, McCormack GH, Vickers JC. Neuro-chemical diversity of dystrophic neurites in the early andlate stages of Alzheimer’s disease. Exp Neurol 1999; 156:100–10

45 Shoji M, Hirai S, Yamaguchi H, Harigaya Y, Kawaraba-yashi T. Amyloid beta-protein precursor accumulates indystrophic neurites of senile plaques in Alzheimer-typedementia. Brain Res 1990; 512: 164–8

46 Walsh DM, Selkoe DJ. Deciphering the molecular basis ofmemory failure in Alzheimer’s disease. Neuron 2004; 44:181–93

47 Hu X, Shi Q, Zhou X, He W, Yi H, Yin X, Gearing M, LeveyA, Yan R. Transgenic mice overexpressing reticulon 3develop neuritic abnormalities. EMBO J 2007; 26:2755–67

48 Stokin GB, Almenar-Queralt A, Gunawardena S, Rod-rigues EM, Falzone T, Kim J, Lillo C, Mount SL, RobertsEA, McGowan E, Williams DS, Goldstein LS. Amyloid pre-cursor protein-induced axonopathies are independentof amyloid-beta peptides. Hum Mol Genet 2008; 17:3474–86

49 Shi Q, Hu X, Prior M, Yan R. The occurrence of aging-dependent reticulon 3 immunoreactive dystrophic neur-ites decreases cognitive function. J Neurosci 2009; 29:5108–15

50 Zheng H, Koo EH. The amyloid precursor protein: beyondamyloid. Mol Neurodegener 2006; 1: 5

51 Cai H, Wang Y, McCarthy D, Wen H, Borchelt DR, PriceDL, Wong PC. BACE1 is the major beta-secretase for gen-eration of Abeta peptides by neurons. Nat Neurosci 2001;4: 233–4

52 Vassar R, Bennett BD, Babu-Khan S, Kahn S, MendiazEA, Denis P, Teplow DB, Ross S, Amarante P, Loeloff R,Luo Y, Fisher S, Fuller J, Edenson S, Lile J, JarosinskiMA, Biere AL, Curran E, Burgess T, Louis JC, Collins F,Treanor J, Rogers G, Citron M. Beta-secretase cleavage ofAlzheimer’s amyloid precursor protein by the trans-membrane aspartic protease BACE. Science 1999; 286:735–41

53 Johnston JA, Ward CL, Kopito RR. Aggresomes: a cellularresponse to misfolded proteins. J Cell Biol 1998; 143:1883–98

54 Waelter S, Boeddrich A, Lurz R, Scherzinger E, Lueder G,Lehrach H, Wanker EE. Accumulation of mutant hunt-ingtin fragments in aggresome-like inclusion bodies as aresult of insufficient protein degradation. Mol Biol Cell2001; 12: 1393–407

55 McNaught KS, Shashidharan P, Perl DP, Jenner P,Olanow CW. Aggresome-related biogenesis of Lewybodies. Eur J Neurosci 2002; 16: 2136–48

56 Diaz-Corrales FJ, Asanuma M, Miyazaki I, Miyoshi K,Ogawa N. Rotenone induces aggregation of gamma-tubulin protein and subsequent disorganization of thecentrosome: relevance to formation of inclusion bodiesand neurodegeneration. Neuroscience 2005; 133:117–35

57 Yi JJ, Ehlers MD. Emerging roles for ubiquitin and proteindegradation in neuronal function. Pharmacol Rev 2007;59: 14–39

58 Cripps D, Thomas SN, Jeng Y, Yang F, Davies P, Yang AJ.Alzheimer disease-specific conformation of hyperphos-phorylated paired helical filament-Tau is polyubiquiti-nated through Lys-48, Lys-11, and Lys-6 ubiquitinconjugation. J Biol Chem 2006; 281: 10825–38

59 Lowe J, Blanchard A, Morrell K, Lennox G, Reynolds L,Billett M, Landon M, Mayer RJ. Ubiquitin is a commonfactor in intermediate filament inclusion bodies of diversetype in man, including those of Parkinson’s disease,Pick’s disease, and Alzheimer’s disease, as well asRosenthal fibres in cerebellar astrocytomas, cytoplasmicbodies in muscle, and mallory bodies in alcoholic liverdisease. J Pathol 1988; 155: 9–15

60 Dickson DW, Wertkin A, Mattiace LA, Fier E, Kress Y,Davies P, Yen SH. Ubiquitin immunoelectron microscopyof dystrophic neurites in cerebellar senile plaques ofAlzheimer’s disease. Acta Neuropathol (Berl) 1990; 79:486–93

61 Gai WP, Blessing WW, Blumbergs PC. Ubiquitin-positivedegenerating neurites in the brainstem in Parkinson’sdisease. Brain 1995; 118 (Pt 6): 1447–59

62 Laser H, Mack TG, Wagner D, Coleman MP. Proteasomeinhibition arrests neurite outgrowth and causes ‘dying-back’ degeneration in primary culture. J Neurosci Res2003; 74: 906–16

Supporting information

Additional Supporting Information may be found in theonline version of this article at the publisher’s web-site:

Figure S1. DAB images of ageing associated dystrophicneurites. Coronal sections of the hippocampus from15-month-old mice were immunostained with the labelledantibodies, which shows the same pattern as the fluores-cence images. Arrowheads indicate clusters of dystrophicneurites. High-power images of clusters are shown next tothe low-power images. Scale bar = 50 mm (Low power).Scale bar = 5 mm (High power).Figure S2. (A) High-power confocal images of clusters ofdystrophic neurites. Coronal sections of the hippocampusfrom 15-month-old mice were immunostained with thelabelled antibodies and confocal images were taken. Scalebar = 5 mm. (B) Coronal sections of the hippocampusfrom 15-month-old mice were immunostained with theAPP-NT antibody. Higher-power image of clusters in thesquare box is shown in the lower panel. Scale bar = 5 mm.(C) Transmission electron microscopy images of clustersof dystrophic neurites. High magnification image of a



dystrophic neurite (arrowhead) filled with translucentor amorphous electrodense materials. Scale bar = 2 mm(upper panel). Scale bar = 500 nm (lower panel).Figure S3. Accumulation of cathepsin-D and g-tubulin inthe dystrophic neurites of lactacystin-treated neurones.Treatment of primary cortical neurones with the protea-some inhibitor lactacystin (10 mM) for 24 h induced accu-mulation of cathepsin-D and g-tubulin (arrowheads) indystrophic neurites, indicating the accumulation of lyso-somes and aggresomes. Scale bar = 20 mm.Figure S4. Accumulation of a-secretase-cleaved APP pep-tides in the dystrophic neurites of lactacystin-treated neu-rones. (A) Immunofluorescence of various APP epitopesin control neurones. (B) After treating with 10 mM lacta-cystin for 24 h, dystrophic neurites that accumulated

cathepsin-D were also immunopositive for 6E10 staining,which detects Ab1–16 (arrowheads); however, immunos-taining for the 4G8 antibody, which detects Ab17–24,was less intense in dystrophic neurites and was notcolocalized with cathepsin-D (arrows). APP-NT immu-nostaining was prominent in dystrophic neurites, butAPP-CT695 immunostaining was light and did not colo-calize with APP-NT (arrowheads), indicating the pre-dominant accumulation of a-secretase-cleaved APPpeptides in the dystrophic neurites of lactacystin-treatedneurones. Scale bar = 20 mm.

Received 9 February 2012Accepted after revision 31 January 2013

Published online Article Accepted on 18 February 2013



α-Secretase cleaved amyloid precursor protein (APP) accumulates in cholinergic dystrophic neurites...

Documents

Transcript of α-Secretase cleaved amyloid precursor protein (APP) accumulates in cholinergic dystrophic neurites...