Jell Inger 2006
Jell Inger 2006
Jell Inger 2006
Kurt A. Jellinger*
Abstract: There are myriads of reasons and ways for a neuron to die, among which apoptosis is a specific form that is
processed in two major signaling pathways, the TNF-receptor-mediated (extrinsic) and the mitochondria-based (intrinsic)
cell death pathway with several avenues of crosstalk between them. The molecular key players of apoptosis, the impor-
tance of the Csp cascade via interaction with different death effector domains and the role of the effector Csp-3 driving the
execution of the cell death program are reviewed. Recent data suggest that caspases converge amyloid and tau Alzheimer
pathologies: β amyloid peptide activates caspases which on turn cleave tau and via phosphorylation of tau initiate tangle
pathology in both Alzheimer disease and other tauopathies. Several mediators show a bifunctional regulation of apoptosis,
with both pro- and anti-apoptotic activities. The latter modify the cell death pathway due to inhibition of Csp activation or
other protective mechanisms and may delay it or, via abortive apoptosis ("abortosis") lead to prolonged survival of nerve
cells. While the role of apoptosis in neurodegeneration is well documented in tissue culture and transgenic animal models,
in human postmorten AD brain its occurrence and role are discussed controversially. Given the short duration required for
the completion of apoptosis and the chronic progressive course of neurodegeneration in Alzheimer disease and related
disorders, the detection of rare neurons displaying morphological signs of apoptosis and expression of the activated key-
executing enzyme Csp-3 is realistic, although there is significantly increased incidence of cells with DNA fragmentaion,
mainly glia, and markers for a "proapoptotic" environment in the aged human brain indicate increased susceptibility of
neurons to metabolic and other noxious factors. Postmortem analysis can bridge some but not all of our knowledge gaps,
but the results are still controversial, and we need a better understanding of the molecular basis and pathways that drive
the yin-yang between neuronal survival and death.
Keywords: Apoptosis, programmed cell death, signaling pathways, caspases, death effector domain, cell cycle, β amyloid pro-
tein, tau phosphorylation, Alzheimer disease
between apoptosis and necrosis [7,13-20], and switching other apoptosis-inducing factors (AIPs), small proteins re-
among the alternative pathways to cell death is relatively leased from damaged mitochondria into the cytoplasm in
common [21-23] as a consequence of the action of calpain early stages of cell death, a family of intracellular cystein
and Caspase families of cystein proteases [23a]. kinases named caspases (cystein aspartyl-specific proteases),
the key executors of neuronal apoptosis [45,46], the Bcl-2
Apoptosis is carried out by an intrinsic suicidal program
and p53 oncogene family, and mitogen-activated protein
of the cell and many factors contribute to the cell destruction
kinase pathways regulated by neurotrophins. Death receptors
pathway, including irradiation, lack of mitochondrial DNA,
are cell surface receptors that transmit apoptotic signals de-
gene expression leading to DNA damage, oxidative stress,
livered from specific ligands [47].They belong to a subgroup
toxins, viruses, withdrawal of neurotrophic support, etc.,
of the tumor necrosis factor receptor (TNFR) superfamily
each demonstrating specificity of function, regulation, and
and are characterized by a so-called death domain (DD) that
pathway involvement [1,8,13-15,24-26].
resides in the cytoplasmic region. Other members are
In the nervous system, apoptosis is necessary for normal TNFRI, Fas, the ectodysplain receptor (EDAR), and the p75
development, but it also occurs in many acute and chronic neurotrophin factor receptor (p75NTR/[48]). Death domain
pathologic conditions, including neurodegenerative and other sequences are all highly homologous and capable of protein-
chronic age-related disorders, such as Alzheimer, Parkinson, protein interactions. They recruit intracellular adaptor pro-
Huntington, and motor neuron diseases [16,19,23,24,27-45]. teins that also contain a death domain, e.g. Fas-associated
The present paper will give an overview of the recent death domain protein (FADD), TNFR-associated death do-
developments in the field of neuronal apoptosis with impact main protein (TRADD), and receptor-interacting protein
on Alzheimer disease (AD) and related disorders. (RIP).
Apaf-1 that has a critical role in apoptotic cell death, is a
KEY PLAYERS IN NEURONAL APOPTOSIS member of the protein family that contains a Csp recruitment
Apoptosis is regulated by an evolutionary conserved domain and regulates apoptosis.
cellular pathway that consists of a number of factors with Previous studies have highlighted the importance of
several possible avenues of crosstalk between them (Fig. 1). caspases for diverse cellular processes in which apoptotic
They are regulated by apoptosis-inducing or death re- Csp activation has been achieved through death receptor
ceptors, e.g. Apo-1/Fas (CD95/Fax receptor), the adaptor signaling.
protein Apaf 1 (apoptotic protease-inducing factor 1) and
Fig. (1). Signal transduction of apoptotic cell death showing the subsequent stages of Csp activation, cleavages of death substrates, and tran-
scriptional regulation of pro- and autapoptotic substances. (mod. from [24,30]). For abbreviations see list at end of text.
Challenges in Neuronal Apoptosis Current Alzheimer Research, 2006, Vol. 3, No. 4 379
Depending on the cell type and the apoptosis stimulus, Two Signaling Pathways of Apoptosis
different caspases are preferentially activated which, in turn,
Caspases which are located in the cytosol as inactive zy-
have different substrate specificities. So far, at least 14 dif-
mogens (procaspases) are activated through two principal
ferent caspases have been identified according to their DNA
pathways, the extrinsic pathway which is initiated by cell
sequence homologue, of which 11 have been found in the
surface death receptor ligation, e.g. cytokin receptors of the
human genome [49,50]. They are divided into upstream ini-
tumor necrosis factor (TNF) family, and the intrinsic path-
tiator caspases with long N-terminal prodomains, and down-
way which arises from the the function of mitochondria (re-
stream effector caspases containing short prodomains
active oxygen species production, membran depolarization,
[45].While initiator caspases, e.g. Csp-1,-8, and -9, can inter-
etc), where release of mitochondrial cytochrome-c (an en-
act with other activating proteins through their long prodo-
zyme of the respiratory chain) into the cytoplasm induces
mains, and activate the short prodomain containing effector
formation of apoptosomes, which leads to activation of Csp-
caspases, e.g. Csp-3, -6,-7,-12, important regulators of post-
9 and subsequently other caspases. There can be a crosstalk
mitotic neuronal homeostasis [45,51,52] Csp-1 is an up-
between these two pathways (Fig. 2). All these pathways are
stream positive regulator of Csp-6-mediated cell death in
associated with activation of Csp-related DNAse (CAD) and
primary human neurons. The activation of Csp-1 must be
accompanied by an apoptotic insult to induce Csp-6-
mediated cell death [52a]. Csp-8, an initiator of the extrinsic
signaling pathway, contains a so-called death effector do-
main (DED) at the N-terminus through which it interacts
with and is activated by other DED proteins. One of these
DED proteins (FADD) additionally contains a so-called
death domain (DD) that has been shown to interact with
death receptors of the TNF family [45]. Although the activa-
tion of Csp-8 in neurons after different death stimuli has
been described [53], the induction of neuronal apoptosis by
death ligands and receptors is still controversially discussed.
Csp-8 activation, by binding of its death effector domain to
the Fas-associated death domain (FADD) initiates the Csp
cascade, via interaction between the death domains of Fas
and FADD. Processing of the key protease, Csp-3, depends
on the activation of Csp -9, and selective peptide inhibitors
of Csp-9 block processing of Csp -3 and -8, and inhibit
apoptosis, wheras a selective inhibitor of Csp-8 blocks nei-
ther processing of Csp-3 nor cell death. Through adaptor
proteins, such as FADD ad TRADO, Csp-8 is autoproteolyti-
cally activated, which, in turn may either cleave Csp-3 di-
rectly or amplify the death signal through translocation of
BID, a mitohondrial protein of the Bcl-2 family. Recent
studies suggest that extracellular signals that lead to oxida-
tive damage and mitochondrial dysfunction with subsequent
cytochrome-c release have the potential to activate the Csp-3
apoptotic pathway [54]. The downstream caspases and their
proteolytic products are recognized markers of apoptosis,
their activation representing an irreversible step in the cell
death cascade, and cells expressing these enzymes are prone
to death. Caspases are contributing to cell death, either as
upstream signals or as downstream effectors contributing to
apoptotic morphology, as well as alternative strategies for
cell death inhibition. Recent studies have revealed that Csp- Fig. (2). Overview of apoptotic signaling through the receptor-
dependent signaling pathways are also linked to cell death by mediated ("extrinsic") and mitochondria-based ("intrinsic") cell
nonapoptotic mechanisms, indicating that apoptosis is not death pathways. The extrinsic pathway involves receptor-mediated
the only form of PCD. These data suggest that canonical activation of Csp-8 with subsequent activation of the effector Csp-
apoptotic pathways include dealth-receptor signaling, control 3, finally leading to DNA fragmentation and apoptotic cell death.
Csp-dependent and –independent cell-death pathways [54a]. Within the intrinsic pathway, Csp-3 is activated after release of
Such alternative strategies may either target catabolic hy- cytochrome-c from mitochondria and formation of the oligomeric
drolases or be aimed at preventing mitochondrial membrane complex called apoptosome consisting of Csp-9, Apaf-1 and cyto-
permeabilization and its upstream triggers.. However, chrome-c. Mitochondrial pathology is controlled not only by Bcl-2
caspases can also contribute to processes that do not culmi- and Bcl-modulating proteins. Multiple crosstalks between the
nate in cell demise [55]. apoptosis pathways and feedback mechanisms exist (mod. from
[29]). For abbreviations see list at end of text.
380 Current Alzheimer Research, 2006, Vol. 3, No. 4 Kurt A. Jellinger
with typical intra-nucleosomal DNA fragmentation (for rev. ships between death and Csp activation: cell death through
see [63]). There is considerable evidence for the relevance of Csp activation, cell death with Csp activation, and cell death
the intrinsic signal transduction pathway for neuronal apop- without Csp activation (the latter two cases being independ-
tosis. DNA damage, increased expression of the tumor sup- ent of caspases). It seems that, in contrast to earlier expecta-
pressor gene p53 or calcium influx by overstimulation of the tions, cell death with Csp activation is frequently not pre-
glutamate receptor (excitotoxicity), damage of components vented by Csp inhibitors, leading to a phenotypic shift in the
of the plasma membrane, formation of free radicals (oxida- morphological manifestation of cell death from apoptosis to
tive stress), and metabolic stress (hypoxia, etc.), can cause non-apoptotic pathways [55], and show a complex interplay
mitochondrial changes resulting in the formation of pores in [64a].
the mitochondrial membrane (permeability transition
Alternatively, the release of the protease AIF is capable
pores/PTP) and release of several apoptosis-relevant mole-
of directly inducing nuclear fragmentation and Csp-3 activa-
cules (cytochrome-c, SMAC/Diablo, AIF). The mitochon-
tion independent of Apaf/Csp-9 (extrinsic pathway) [65].
drial PTP appears to be crucial in both necrotic and apoptotic
BAX, a pro-apoptotic homologue bound to mitochondria,
cell death [55a]. Its activation increases the inner mitochon-
which also includes release of cytochrome-c by heterodi-
drial membrane permeability to solutes with a molecular merizing with Bcl-XL and displaying it from the inacctive
mass up to 1.5kDa. The inner mitochondrial membrane ade- Apaf-1/Csp-9, activates Csp-3, which cleaves other caspases
nine nucleotide transporter, the voltage-dependent anion in the death-cascade [43,66,67]. It further promotes neuronal
channel and the peripheral GABA receptor in the outer
cell death in response to cytotoxic injury with a key role for
membrane and mitochondrial creatinine kinase Bax interact p53 activation and the Bcl-2 family [68]. Conversely, anti-
with the voltage dependent anion channel to accelerate
apoptotic members of the Bcl-2 family, e.g. Bcl-2, may in-
opening of the PTP which is favord by Ca2+ and oxidizing
hibit the opening of the mitochondrial permeatbility transi-
agents, contributing the cytochrome-c release. Neuronal Ca 2+
tion pores (MPT) [69] and inhibit both apoptosis and necro-
overload, such as follows extensiv stimulation of Ca2+ per-
sis, presumably by interfering with reactive oxygen mole-
manent excitatory amino acid receptors, can cause cell death.
cules, while Bcl-XL acts by holding the pro-apoptotic Apaf-
Recent evidence suggests that the accumulation of Ca2+ into 1/Csp-9 complex inactively bound to the mitochondrial
mitochondria during episodes of cellular Ca2+ overload indi-
membrane [45,50]. Among the Bcl-2 binding proteins with
cates a cascade of events that culminate in cell death. The
anti-apoptotic activity is the BAG family [70], having potent
latter appears to require not only mitochondrial Ca2+ over-
neuroprotective action by binding to heat shock 70 molecular
load, but rather a combination of raised intramitochondrial
chaperons and linking cellular stress responses to the apop-
Ca2+ concentration with increased production of nitric oxide
totic cell program [71]. Another Bcl-2 binding protein with
and possibly other oxyradical species. Cell death may pro-
neuroprotective activity is the bifunctional apoptosis regula-
ceed through either necrotic or apoptotic mechanisms, de-
tor BAR, a multidomain protein that is capable to inhibit
pending on the rate of consumption and depletion of ATP apoptosis induced by TNF family death receptors (extrinsic
[56]. Mitochondrial dysfunction leading to cell death are pathway) as well as mitochondria-dependent apoptosis.
controlled and mediated by proteins of the proto-oncogene
Moreover, BAR has been suggested to act as domain similat
Bcl-2 family, which includes a highly homogeneous group of
to classical death effector domain ("pseudo"-DED) that me-
mitochondrial proteins with pro-apoptotic (enhancing) ef-
diate Csp-8 binding. Therefore, BAR has been suggested to
fects (e.g. Bax, Bad, Bid, Bak) and others with anti-apoptotic
act as a scaffold protein that can bridge components of both
(preventive) effect (e.g. Bcl-2, Bcl-XL, Mcl-1). Formation of
apoptotic pathways, and is considered a bifunctional apopto-
homotypic or heterotypic dimers which affect the formation
sis regulator. It is highly expressed in neurons and promotes
of a permeability transition zone in the mitochondrial mem- survival after diverse apoptotic stimuli [72]. Cells overex-
brane and release of cytochromce-c, as well as the equilib- pressing both Bcl-2 and Bax show no signs of Csp activation
rium shift between pro- and anti-apoptotic members of the
and survive even with significant amounts of cytochrome-c
Bcl-2 family may determine the sensitivity of a cell to
in the cytoplasm, indicating that Bc-2 can prevent Bax-
apoptotic stimuli [27,50,57-59]. Following its release from
induced apoptosis by other mechanisms [73]. Thus, apopto-
the mitochondrial intermembrane space into the cytosol, cy-
sis can be delayed by members of the Bcl-family that may
tochrome-c leads to the formation of oligomeric cytoplamic
elicit protective effects by preventing the release of
complexes (apoptosomes). This results in Csp-9 activation
Apaf1/Csp-3 from mitochondria [50]. The role of the
and further activation of downstream Csp-3 [60]. The inter-
proteasome in neuronal apoptosis is poorly understood since
action of Apaf-1 and Csp-9 is mediated through a CARD both anti- and pro-apoptotic effects result from proteasome
(Csp-associated recruitment domain) domain contained in
inhibition, which exerts an acute pro-survival action by
both proteins [61]. For the activation of Csp-9, a mechanism
stabilizing MEF2 transcription factors. However, chronic
called "induced proximity" may be responsible [62], and the
proteasome inhibition causes a build-up of phosphorylated c-
promotion of multiple Csp molecules promotes cleavage and Jun and Bim, which eventually overwhelms the effects of
transition into the fully active form. Actived Csp-9 cleaves MEF2 and triggers apoptosis [73a].
and activates the effector Csp-3, which is responsible for
driving execution of the cell death program [50,63]. When Other molecular changes associated with Csp-
Csp activation is inhibited at or downstream of the apopto- independent pathways of apoptosis include induction of
some, neurons undergo a delayed, Csp-independent death, proapoptotic proteins, such as prostate apoptosis-response-4
for which alterations of mitochondria and non-Csp proteases (Par-4) [74] and certain members of the pro-apoptotic Bc-2
plays a significant role [64]. One can envisage three relation- family (see above).
Challenges in Neuronal Apoptosis Current Alzheimer Research, 2006, Vol. 3, No. 4 381
The p53 family of oncogenes are transcription factors Poly(ADP-ribose)polymerase-1 (PARP-1) is a nuclear
playing a major role in determining the cell-fate in response enzyme, activated by DNA strand breaks to partizipate in
to DNA damage. P53 induces the pro-apoptotic proteins DNA repair. Its overactivation by cellular insults depletes its
Fas/Apo1, DRS, Bax, and other cell death-related proteins substrate NAD(+) and then ATP, leading to an energy deficit
and represses survival mediators, such as IGF-15. In addition and cell deth [85,86]. This mechanism appears to be promi-
to such transcriptional control of the cell death machinery, nent in ischemia and neurodegenerative processes. However,
p53 may more directly trigger apoptosis by acting at the PARP-1 activation appears to be predominantly necrotic,
level of mitochondria, a process that can occur in synapses while apoptosis is associated with PARP-1 cleavage, which
(synaptic apoptosis [75]). Adult motor neuron apoptosis is may conserve energy needed for the apoptotic process.
mediated by nitric oxide and Fas death reception linked
Because of the genetically driven nature of apoptosis,
DNA damage and p53 activation [76]. P73, a homologue of
nuclear transcription factors, specifically, immediate early
p53, is able to induce apoptosis in several cell types. In con-
genes (IEG) can influence the initiation and execution of this
trast to p53, the Tap73 gene has two distinct promoters cod-
cascade. The proto-oncogenes c-Jun and c-Fox, members of
ing for two protein isoforms with opposite effects: while the
the IEG family, have a putative role in transcriptional regu-
transactivation proficient Tap73 shows pro-apoptotic effects, lation of apoptotic genes, e.g. p53 [87]. C-Jun is expressed in
inducing the mitochondrial pathway by directly transacti- extended periods, whereas c-Fos appears to be a late event in
vating both Bax and the BH3 only protein PUMA promoter, apoptosis, occurring just before chromatin condensation.
which is strong and lethal, while the amino-terminal-deleted
IEG-encoded proteins, c-Jun/Ap1 and Aβ-2 also appear late
Udelta Np73 has an anti-apoptotic function. The balance in the process of apoptosis, in particular, following excitox-
between these two isoforms finally regulates cellular sensi-
icity [88]. Other anti-apoptotic signals are mediated by c-Jun
tivity to death signals [77]. calpain-induced activation of
amino terminal kinase /JNK [89] or the activation of tran-
Cdk5 (cyclin-dependent kinase 5)/p25 that is a general fea-
scription factors, such as CREB (cAMP-responsive element
ture of cell death can occur in the absence of p35, Apaf-1,
binding protein) and NF-κB (see [27,50,90]. It is still
Csp-9 and -3, indicating that its activation relates more to
unknown whether oxidative-stress-induced neuronal PCD
cell death than to a specific pathway of apoptosis [78]. requires re-initiation of the cell cycle [90a], which in dying
Analysis of PCD in neuronal demise has led to identification
neurons could be a tentative attempt to repair oxidative-
of several associated phenomena, such as re-initiation of the
stress-induced DNA damage, and might thus be considered
cell cycle and the key role of oxidative stress, although
as a physiological 'defence' mechanism in the presence of
putative causal relationships between these events are still
damaged DNA [90b].
debatable [78a].
Increasing evidence suggests that the regulation of neu-
DNA Repair and Cell Death ronal cell death is complex, utilizing multiple pathways that
are depending on the neurodamaging insult and are influ-
DNA repair is suggested to be compromised in AD [79]. enced by subtle differences among cell phenotypes.
DNA double strand breaks are one serious form of DNA
damage. Left unrepaired, they can lead to cell death. The APOPTOSIS IN NEURODEGENERATIVE DISEASES
cause and natural function of such breaks remains a mystery,
but there is now renewed speculation that DNA recombina- Although the various molecular players in apoptotic cell
tion might be occurring during neuronal development death are becoming increasingly well known, whether or not
[80,81]. they are involved or may interact with one another in the
neuronal cell cascade, in various neurodegenerative proc-
Among the family of apoptosis suppressors called IAPs esses. Nature, time course and molecular causes of cell death
(inhibitory of apoptosis proteins) with a BIR (buculovirus in neurodegenerative disease and their relations to basic
IAP repeat domain) identified in humans [50], at least one is
processes discussed above are still a matter of discussion.
expressed in neurons (neuronal apoptotic inhibitory pro-
Similar death-signalling pathways might be activated in neu-
tein/NAIP). IAPs prevent the "unintentional" activation of
rodegenerative disordes by abnormal subcellular protein
effector caspases, but are negatively regulated by the mito-
structures and cytoplasmic or nuclear fibrillary inclusions,
chondrial factor SMAC/Diablo. It is released with cyto-
such as amyloid β peptide (Aβ), tau proteins, parkin, hunt-
chrome-c, by which the effective operational sequence of the
ingtin, α-synuclein, etc. [31,41,91-96]. Alterations in cellular
apoptotic cascade after approriate stimulation is ensured,
homeostasis that affect protein folding in the endoplasmic
allowing Csp-9 and -3 to be activated through their interac- reticulum trigger a signaling pathway known as the unfolded
tion with Apaf-1 [82]. Other endogenous inhibitors of the
protein response (UPR). The initially cytoprotective UPR
apoptotic protein family (IAP), e.g. NIAP, X-liked inhibitor
will trigger an apoptotic cascade if the cellular insult is not
of apoptosis protein (X-IAP), survivin, c-IAP-1 and 2, and
corrected; however, the proteins required to initiate this cell
viral antiapoptotic proteins modulate Csp activity in various death pathway are poorly understood [96a].
points within these pathways. Head shock protein 70 has
been added to the list of molecules with strong anti-apoptotic Implication of apoptosis as a general mechanism in many
properties [83]. However, blockage of the Csp executive neurodegenerative disorders has largely been supported by
pathway may only temporalily rescue damaged neurons, evidence from animal models and tissue cultures [see 96b],
which does not necessarily warrant cell survival, and classi- while studies on human postmortem brain tissue have
cal apoptosis features can still appear in Csp-inhibited neu- yielded contradictory results, because clear detection of
rons [84]. In some cases, proteolysis activates Csp substrates, apoptotic cells is difficult or problematic. This may be ex-
in others it inactivates or destroys them. plained by the fact that cell death in these chronic disorders
382 Current Alzheimer Research, 2006, Vol. 3, No. 4 Kurt A. Jellinger
occurs over decades, while the suicidal program in the single herited mutations of genes encoding APP and presenilins in
cell is executed within a few hours . Thus, synchronous de- rare familial EOAD, whereas in LOAD there are compelling
tection of a substantial number of apoptotic neurons at any reasons that a failure of degradation or clearance of Aβ from
given time point appears almost impossible, although some the brain underlies its accumulation (see [104]). NFTs con-
authors based on currently available methods for detection of stituted of phosphorylated and aggregated forms of the mi-
apoptosis, i.e. the presence of morphological markers crotubule-binding tau proteins that stabilize the cytoskeleton
(TUNEL/ISEL staining associated with the expression of and partizipate in vesicular and axonal transport [105]. Pre-
ARPs, oncogenes, and other detectable players involved in viously, it was hypothesized that the development of the
the cascade of PDC), have reported variable incidence of classical hallmarks of AD - Aβ and tau - occur independ-
apoptotic cell death in a variety of neurodegenerative dis- ently, whereas recent data indicate nonoverlapping but syn-
eases (for rev. see. [24,30,31,97]). In many of these condi- ergetic action of both pathologies in sporadic AD [106].
tions, mitochondrial dysfunction and the resulting ATP de-
pletion may preclude Csp activation, and consequently Caspases Converge Amyloid and Tau Pathologies
switch execution of cell death towards necrosis. A block or
partial inhibition of the typical apoptotic demise may have According to the amyloid cascade hypothesis [107,108],
Aβ accumulation in the brain derived from imbalance be-
profound implications in vivo, as persistence within the
tween Aβ production and clearance is the primary influence
nervous system of damaged, but "undead" cells, followed by
and APP is driving tau pathology with formation of NFTs.
delayed lysis may favour neuroinflammatory reactions. Fur-
This is supported by tau co-localization with Aβ-42 and the
thermore, caspases may be involved in loss of neurons, but
not in the loss of connectivity that seems to initiate degen- induction of PHF-like filaments in vitro [109].Both exposure
erative processes in the nervous system. Some recent find- to Aβ and tau lesions induce tissue damage, their interrela-
ings suggest that degeneration of neurons may use multiple tionship being under discussion. Whereas previously it was
suggested that Aβ-induced neuronal death is Csp-
executive pathways [22,24,41].
independent [110], other data indicate that multiple caspases
are involved in Aβ-induced neuronal cell death [111] and
NEURONAL CELL DEATH IN ALZHEIMER DIS-
that mitochondrial dysfunction and endoplasmic reticulum
EASE
failure play an essential role in Aβ-induced apoptosis neu-
General Neuropathology ronal death [112]. It has been demonstrated that Aβ induces
cell death by an apoptotic process related to Csp-3 activation
Alzheimer disease (AD), now recognized as the most
that is regulated by calcineurin-mediated BAD dephos-
common cause of dementia in the elderly, shows increasing
phorylation - suggesting to represent upstream premitochon-
prevalence and incidence with advancing age in an exponen-
drial signaling events [113]. Recent data indicate a link be-
tial fashion. Its prevalence increases from 1% in the seventh
tween the development of Aβ and NFT/tau pathologies may
decade, with doubling every subsequent 5 years, to up to
be activation of caspases by Aβ which in turn cleaves tau
50% over age 85 years. The incidence exceeds 1/100 per- and may initiate or accelerate the development of tangle pa-
sons/year from age 70 to 80, and reaches 2/100/year until age
thology [39]. Csp cleavage of APP and tau has been demon-
90, now affecting 20 to 35 million people worldwide [98].
strated in AD [28,114-116]. Aβ accumulation, that is en-
Neuronal and synaptic loss occurs during the progress of
hanced by molecular chaperons, including metals like copper
AD, causing cerebral volume loss (atrophy) and progressive
and proteins that interact with transducers of the Wnt/β cat-
disruption/disconnection of neuronal circuits as a major sub-
enin signaling pathway [117], triggers Csp activation which,
strate of dementia. The histological hallmarks of AD are
in turn, leads to Csp-cleavage of tau that is an early event,
extracellular deposition of β amyloid protein (Aβ) in plaques
which may precede hyperphosphorylation in the evolution of
and cerebral vasculature (cerebral amyloid anigopathy/CAA) AD tangle pathology [32,115,118,119]. In addition, Aβ may
and accumulation of hyperphosphorylated tau proteins
disrupt the proteasomal degradation of tau [120] and acti-
forming paired helical filaments (PHF) in neurons (neurofi- vates caspases that can cleave tau [32,118,121]. Tau pathol-
brillary tangles/NFT), dendrites (neuropil threads), and
ogy in transgenic (tg) mice is exacerbated by Aβ [122,123].
around plaques (neuritic plaques). Although most of these In triple-tg mice (3xtg-AD), Aβ not only preceeded but ac-
changes are nonspecific, the diagnosis of AD depends on
celerated the development of tau pathology [123,124]. Cur-
their semiquantitative assessment.or topographic staging [99- rently, the mechanisms, by which Aβ drives tangle formation
101]. Other lesions are microglial activation (inflammatory are unresolved. However, APP and Aβ activate kinases im-
cascades), astroglial proliferation, granulovacuolar degen-
plicated in tau phosphorylation, including glycogen synthase
eration, and Lewy bodies (for rev. see [102,103]).
kinase-3β (GSK-3β), mitogen-activated protein kinase
The pathogenesis of AD is unclear. Although it is widely (MAPK), and cyclin dependent kinase 5 (Cdk5), stress
believed that Aβ is seminal, little is known about the rela- kinases SAPK/JNK and other kinases [120,124-126]. Acti-
tionship between APP-Aβ mismetabolism and tau changes in vation of c-Jun N-terminal kinase (JNK) and p38 kinase, two
sporadic AD, where the pathogenic trigger has not been members of the MAP kinase family, that are strongly in-
identified. Derived from a large transmembrane glycoprotein duced by various stresses, including oxidative stress and
precursor, amyloid precursor protein (APP), Aβ is released have been involved in regulation of apoptosis, are also in-
as soluble peptides, with Aβ-40 and Aβ-42 as the most volved in tau phosphorylation [127]. It is modulated by the
abundantly produced isoforms. The balance between bio- sequential and cumulative action of distinct caspases which
genesis versus catabolism of Aβ is disturbed through en- are recruited in PHFs and other tau containing structures, e.g.
hancement of β- and gamma-secretase activity by some in- Pick bodies. Detection of granular precipitation of other
Challenges in Neuronal Apoptosis Current Alzheimer Research, 2006, Vol. 3, No. 4 383
phosphorylated MAP kinases (ERK 1 and 2) associated with APP, presenilins (PS), ApoE and tau can cause impairment
early tau deposition in AD and other tauopathies, suggesting of axonal transport of APP, suggesting a possible link be-
activated Ras in the upstream activation of the phosphoryla- tween the key proteins in AD [143,144], leading to axonal
tion-independent MEK1/ERK pathway of tau, however, was depletion of critical components and neurodegeneration
not associated with increased nuclear DNA fragmentation [145,146].
and cell death [128,129].
Delta tau occurs early in the development of tangle pa- Molecular Pathways in AD
thology within AD brain, in tg-mouse AD models, but also The molecular causes of AD are unresolved, but a variety
in other neurodegenerative diseases associated with tau and of pathogenic factors are implicated in the complex cascade
α-synuclein pathology, like Pick disease, progressive su- of events leading to neuronal degeneration: a) reduced cere-
pranuclear palsy (PSP), corticobasal degeneration (CBD), bral energy and glucose metabolism related to disorders of
but not in frontotemporal dementia (FTD), suggesting that the insulin receptor transduction pathways, b) oxidative
Csp-cleavage and activation of certain kinases play an im- stress with accumulation of cytotoxic free superoxide radi-
portant role in the development of tau and other pathologies cals and advanced glycation products (AGE), lipid peroxida-
[31,124]. tion products, and abnormally oxidated cytoskeletal proteins;
Alterations in tau phosphorylation and tau cleavage by c) dysfunction of mitochondria leading to a cellular bioener-
caspases have been reported in neuronal apoptosis. The getic crisis that may include cell death; d) misfolding and
cleavage produced tau generated by Csp-3 is pro-apoptotic accumulation of insoluble fibrillated proteins due to disor-
[119,121,130,131], and Csp inhibition prevents tau cleavage ders of the proteasome and lysosomal systems with forma-
without reversing changes in the tau phosphorylation linked tion of inclusions (NFTs, LBs), e) trace elements, such as
to apoptosis [132]. In AD brain, Csp-cleaved tau co-localizes iron may serve as catalysers for free radical formation, lead-
with both intracellular Aβ and activated Csp-3, mainly in ing to oxidative modification of tau and Aβ, increased oxi-
tangle-bearing neurons [32,133]. Increased expression of dative damage associated with ApoE ε4, f) reduced defense
Bcl-2 and activated Csp-3 have been observed in AD neu- enzyme activity, g) antioxidant activity, such as upregulation
rons and glia showing DNA fragmentation with negative of the enzyme hem oxygenase-2 in tangle-bearing neurons,
correlation to glutamate receptors, but related to spliced APP h) impaired calcium homeostis, i) decrease in the endoge-
forms, indicating that they may lead to activation of Csp-3 nous levels of neurotrophic factors that increase the suscep-
that defines the late stages in the cell death cascade [134]. tibility of neurons to oxidative stress and may lower synapse
Activated Csp-6 in pretangles and in the limbic cortex are density. In addition, these factors contribute to both abnor-
seen in in early stages of AD [135], and other studies dem- mal phosphorylation of cytoskeletal tau protein and prote-
onstrated Csp activation in mild cognitive impairment olytic processing of APP by presenilins with increased pro-
(MCI), but not AD subjects [131], although Csp gene ex- duction of Aβ-42, the main component of senile plaques,
pression in the brain has been considered a function of the that, via, excess glutamine production and other excitatory
clinical progress of AD [136]. While Csp-6-cleaved tau was toxins, may lead to neurodegeneration. While Aβ-42
found in neuritic plaques, neuronal threads and NFTs Csp-6 aggregates very rapidly to form early diffuse plaques,
itself primarily is located in neurites [119]. This suggests that supplementary Aβ-40 deposition is required to form mature
apoptosis-like mechanisms can damage synapses, axons, and neuritic plaques and CAA. Aβ-40 further activates the NF-
dendrites without causing overt neuronal death [39]. These κB pathway, the inhibition of which has been known to
results suggest that the activation of apoptosis-like mecha- prevent Aβ-induced release of cytochrome c, restore the
nisms may be involved in AD pathogenesis [137], and that levels of Bcl-X, and to prevent intraneuronal accumulation
PTH-tau is cleaved by Csp-3 during apoptosis and neurode- and secretion of Aβ-42, thus being a possible target for
generation in AD. The microtubule-depolymerizing agent preventing Aβ-induced neuronal death [146a]. Phospholipid
colchicine induces tau dephosphorylation and Csp- breakdown of neuronal membranes and Aβ deposition, both
independent tau cleavage and degradation that are blocked exaggerated by ApoE ε4, activate microglia secreting cyto-
by inhibiting protein phosphatase 2A (PP2A) by odadaic acid kines, acute phase reactants, nitric oxide and other sub-
[138]. Thus, Csp cleavage of tau provides a mechanistic link stances mediating inflammatory reaction (for rev. see [147-
between the development of Aβ and NFTs and includes con- 149]). The pathogenic cascade of AD finally resulting in
vergence pathologies in AD [39]. However, it remains to be neuronal cell and synapse loss exceeding 50% in certain
determined with certainty if the proteolytic events responsi- brain regions, is suggested to be initiated decades prior to
ble for conformation changes of tau and other proteins are onset of clinical symptoms and manifest AD is considered a
necessary and early events that lead to neurodegeneration. late stage of these processes [106,150]. However, little is
Alternatively, they could be secondary processes in degener- known about the molecular course of neuronal degeneration
ating neurons subjected to activation of apoptotic mechanism and its relationship to plaques and tangles.
on flooding of intracellular components by calcium due to
other underdetermined pathologic processes [139]. Csp acti- Challenges of Neuronal Apoptosis in AD
vation has been shown to inhibit proteosomal degradation of Despite demonstration of DNA fragmentation and an
cellular substrates. so facilitating the execution of the apop- upregulation of pro-apoptotic and cell death regulation pro-
totic program by providing a feed-forward amplification loop teins, it is still unclear whether the apoptotic process is di-
[140]. Others suggest that tau is essential for Aβ-induced rectly responsible for neuronal cell death in AD (see
neurotoxicity and may precede the occurrence of Aβ depos- [24,40]). Cell death is preceded by the activation of caspases
its [141,142]. Misregulated APP and tau by abnormalities of and altered expression of the Bcl-2 family proteins and other
384 Current Alzheimer Research, 2006, Vol. 3, No. 4 Kurt A. Jellinger
ARPs [24,67]. In familial AD, mutated genes alter the prote- of protein kinetics, in both granulovacuolar bodies nd NFT-
olytic cleavage of APP. with simultaneous intracellular depo- bearing neurons suggests a link between these two lesions
sition of a neurotoxic form of Aβ-42 suggesting that this is [170].
an early event in AD, sensitizing neuronal cells to apoptotic
Increased expression of pro-apoptotic proteins or gene
stimuli in vitro [121]. Malfunctioning of cyclin-dependent
products (c-Jun, c-Fos, Par-1, Bax, Bas, APO1-/Fax, etc) but
kinase 5 (Cdk5) through aberrant proteolytic cleavage of its also of anti-apoptotic Bcl-2 family proteins and of repair
neuronal activators p35 and p39 is involved in neurodegen- enzymes such as redox factor 1 (Ref-1), a potent activator of
eranon in AD. Recent data are indicative for a role of the p53, have been observed in AD brain [24,116,119,
Cdk5 molecular complex in the genetic etiology of early- 135,169,171-174], but there is a disturbed balance between
onset AD, and suggest that a yet unknown functional variant
pro- and anti-apoptotic proteins which is associated with
in Cdk5 or in a nearby gene might lead to increased suscep-
incomplete cell cycle activation in postmitotic neurons pos-
tibility for early-onset AD [151]. In APP and PS-1 double-
sibly leading to their elimination via apoptosis [175], while
transgenic (tg) mice, intraneuronal Aβ accumulation pre-
DNA replication precedes neuronal death in AD [176]. In
cedes plaque formation, However, intraneuronal Aβ is not
some cells showing DNA damage, co-expression of ARPs,
detected in brains of aged double-tg mice [152], and results like c-Jun, Bax and Bcl-2 with decreased levels of Bcl-2 in
from these tg mice remain contradictory with respect to the tangle-bearing neurons have been reported [169,172]. Im-
functional relevance of ARP proteins in human AD. Activa- munocytochemistry shows that a significant percentage of
tion of p53-SAPK/JNK are believed to be necessary for Aβ-
hippocampal neurons in mild cognitive impairment (MCI)
induced cell death in vitro [70,153-160]. However, double- express three cell-cyclerelated proteins, proliferating cell
immunohistochemical studies have shown the lack of co-
nuclear antigen, cyclin D and cyclin B [177]. The percentage
localization of both proteins, and of cleaved Csp-3 expres-
is similar to that found in AD but signifianctly higher than in
sion in AD and other tauopathies [161,162]. It seems feasible
normal controls, while in entorhinal cortex, the density of
that oxidative stress in AD leads neurons to elicit adaptative
cell cycle-positive neurons was greater in MCI than in AD.
responses, as the activation of stress-activated kinases that
These findings support the hypothesis that both the mecha-
partizipate in tau phosphorylation but do not mediate cell nism of cell-cycle induced death and the rate of cell loss
death. Several survival signals may converge in phosphory-
(slow atrophy over months) are identical in all stages of the
lated tau-containing neurons [163]. Nevertheless, the altered
AD disease process. Despite large numbers of cells with
degradation and pathological aggregation of mutated pro-
DNA fragmentation in the hippocampus, only exceptional
teins, with consecutive activation of the apoptotic machinery,
neurons display the morphology of apoptosis or show diffuse
seem to be crucial for neuronal death in AD.
expression of ARPs, although the intensity of Bcl-2, c-Jun,
Human postmortem AD brain compared to age-matched Apo-1/Fas, and several stress proteins shows no differences
controls showed a 50-fold increase of DNA fragmentation in between AD and control brains [178]. Other groups have
neurons and 35-fold in glial cells, microglia and oligoden- reported similar levels of BAX, Bcl-2, and TRAAD (TNF
droglia being most affected, whereas only 25% of all degen- receptor-associated death domains) in AD and age-matched
erating cells represent neurons [164]. Most of the TUNEL- control brains [166,173,175]. Despite large numbers of cells
positive neurons are seen in the temporal allocortex, the re- with DNA fragmentation particularly in the severely dam-
gion initially and most severely involved in AD. Only 13- aged hippocampus, only in 0.02-0.05% of hippocampal neu-
50% (mean 28%) of degenerating neurons are located in or rons apoptotic featurs and expression of activated Csp-3
near Aβ deposits, but these are almost 5-fold more frequent have been observed (Fig. 3a,b), indicating the rare occur-
than neurons without contact to plaques. NFTs involved rence of apoptotic cell death in AD [179]. In AD brain,
around 40% (range 18-66%) of all degenerating neurons, TNFR1 has shown increased expression and is related to the
which means an approximately threefold increased risk of apoptotic process, whereas down-regulation of DENN-
degeneration compared to tangle-free neurons. Other authors MADD, an FNF receptor binding protein, correlates with
found nuclei adjacent to Aβ plaques to be almost always neuronal cell death in AD brains and hippocampal neurons
non-apoptotic, and reported a significant negative correlation [180].
between the average number of NFTs in neocortex and the Recent evidence suggests that two intrinsic pathways,
percentage of apoptosis-like nuclei [165-168]. These data
mitochondrial dysfunction and endoplasmic reticulum stress,
suggest that both AD lesions are not necessarily associated are central in the execution of apoptosis in AD, mediated by
with cell death because it frequently occurs at a distance organelle dysfunction [181]. In particular the cleavage of
from plaques or in neurons without NFTs [164]. However, cytoskeletal components, which clearly alters cell morphol-
co-localization of activated Csp-3 with 30% of senile
ogy, may either activate cell death signals or disrupt survival
plaques suggests that apoptosis may contribute to cell death
signals necessary to suppress cell death[182]. Accumulation
resulting from amyloidosis and plaque formation [28]. On of both Aβ-42 and p53 was found in degenerating neurons in
the other hand, activated Csp-3 was detected in more than both tg mice and human AD brain. Thus, the intracellular
50% of hippocampal neurons showing granulovacuolar de-
Aβ-42/p53 pathway may be directly relevant to neuronal loss
generation (Fig. 3d) in AD; it was restricted to the granules
in AD. Although neurotoxicity of extracellular Aβ is well
which often showed co-localization with tau protein, sug- known and synaptic/mitochondrial dysfunction by intracel-
gesting a "pretangle" stage (Fig. 3c,e) but no morphological lular Aβ-42 has recently been reported, intracellular Aβ-42
signs of apoptosis [161,169]. Although neurons with com-
may cause p53-dependent neuronal apoptosis through acti-
pleted NFTs usually do not express activated Csp-3, the
vation of the p53-promoter, thus demonstrating an alterna-
demonstration of elevated casein kinase 1 (CK-1), a member tive pathogensis of AD [183].
Challenges in Neuronal Apoptosis Current Alzheimer Research, 2006, Vol. 3, No. 4 385
Fig. (3). (a) Activated Csp-3 in two apoptotic hippocampal neurons with condensed nucleus and shrunken cytoplasm but without neurofi-
brillary tangles. (b) Hippocampal CA1 neuron in Alzheimer disease (Braak stage 5) immunoreactive for activated Csp-3 (brown-yellow) with
condensed framgented nucleus and shrunken cytoplasm (arrow). Other neurons show no nuclear changes and are negative for CM-1 anibody
(x 1000). (c) Neurons in subiculum of AD brain. One neuron (arrowoheads) shows granulovacuolar degeneration immunolabeled with CM-1
antibody marking activated Csp-3 (brown), while other cytoplasmic compartments remain unstained for CM-1 but slightly express AT-8
antibody ( blue), suggesting finely granular accumulation of hyperphosphorylated tau (pretangle). Two other neurons with neurofibraillary
tangles express AT-8 (blue) but are negative for CM-1; one unaffected neuron (left bottom ) is unlabeled for both antibodies (x 1000). (d)
Hippocampal neurons in AD brain showing multiple granulovacuolar degeneration that is immunoreactive with the CM-1 antibody against
activated Csp-3 (x 500). (e) Hippocampal CA 1 neuron in AD brain double labeled with AT-8 showing a small tangle (arrowhead, blue) and
diffuse AT-9 positivity in a neuron with CM-1 positive granulovacuolar degeneration (arrow, brown) (x 1000).
The detection of active caspases and the accumulation of hand, the significantly increased incidence of cells with
cleaved substances, such as fodrin, actin and APP, in human DNA fragmentation together with a "proapoptotic" environ-
postmortem AD brain tissue supports the hypothesis that ment in the aged brain indicate increased susceptibility of
apoptotic -like mechanisms rarely contribute to neuronal loss AD neurons to metabolic and other noxious factors [178].
in AD [40,114,115]. Furthermore, new data support the hy- Incomplete cell cycle activation in postmitotic neurons pos-
pothesis that Csp-mediated cleavage of critical proteins con- sibly may lead to their elimination via apoptosis, while a lack
tributes to neuronal degeneration [119,137]. However, of effective apoptotic signal propagation to downstream Csp
modulation of distal substrates of activated Csp-3 may lead effects to abortive apoptosis or "abortosis" [44], may lead to
to further modifications of the cell death pathway and may prolonged survival in particular of neurons with NFTs [186],
explain absent evidence demonstrating morphological signs with autophagy as possible mechanism in early stages of
of apoptosis in AD susceptible neurons [169], and the lack of PCD [12,16]. The fact that early- and late-stage NFTs are not
some downstream effector caspases in AD argues against a co-localized with apoptosis-like nuclei [16,166,179], in
specific role of the apoptotic cascade [16,24,41]. The conjunction with reports that most neocortical NFTs in AD
Fas/Fas-1 signaling system also does not appear to be in- are intracellular [187,188], and recent evidence that NFTs in
volved in AD neurodegeneration [129], while mitochondrial experimental models do not directly correlate with neuronal
DNA-damage may be a major mechanisms of cell loss loss [189,190], suggest that affected neurons may be able to
[57,184]. withstand NFT formation for long periods of time before
frank degeneration. Thus, AD represents the first in vivo
The detection of rare hippocampal neurons, displaying
the morphologic signs of apoptotis and expression of acti- situation reported in which the initiation of apoptosis may
not directly lead to cell death. In those surviving neurons it
vated Csp-3, appears fairly realistic given the short duration
required for the completion of apoptosis and the protracted can be suggested that viability is, in part, maintained by the
lack of distal transmission of the Csp-mediated apoptotic
course of neurodegeneration in AD [72,185]. On the other
386 Current Alzheimer Research, 2006, Vol. 3, No. 4 Kurt A. Jellinger
signals. The biochemical cell death pathways that lead to the BIR = Baculovirus IAP receptor domain
morphotype called apoptosis ("falling of") are many. As of C = Cytochrome c
yet, there is no simple type of evidence that this by itself will
form the sufficient and necessary criteria that lead to the CAD = Csp-associated DNase
conclusion of cell death by a particular apoptotic mechanism. CAPDH = Glyceraldehyde-3-phosphate dehydrogenase
However, if we do not conflate mechanisms with end-stage
histopathological features, we think that confusion can be CARD = Csp-associated recruitment domain
diminished as well as novel environments such as AD better Cdk5 = Cyclase-dependent kinase 5
understood, when our evidence argues for neurons trying to
avoid apoptosis. CK-1 = Creatin kinase 1
CREB = cAMP-responsive-element binding protein
CONCLUSIONS
Csp = Caspase
There are many reasons and myriad ways for a neuron to
DAC = Death-inducing signaling complex
die. The reasons range from the good (developmental prun-
ing) to the bad (succumbing to insults such as oxidative dATP = Deoxyadenosine triphosphate
stress, amyloid or other misfolded proteins). The pathways
DD = Death domain
that have been studied overlap at many levels, and staying on
top of this developing way of interactions is a challenge. DED = Death effector domain
Although protein misfolding/misassembly and aggregation EDAR = Ectodysplasin-A receptor
currently appear to be critical and early suspects in the
pathogenesis of AD and other neurodegenerative disorders, Endo-6 = Endonuclease 6
one may also suggest that apoptosis contributes to neuronal EOAD = Early onset AD
death in several ways. On the basis of currently available
data it has to be stated that, although many in vitro and in FADD = FAS-associated death domain protein
vivo studies favor apoptosis in AD and other neurodegenera- IAP = Inhibitor of apoptosis protein
tive disorders, the majority of human brain tissue studies
have yielded controversial results, and there is increasing ICAD = Inhibitor of CAD
evidence for alternative mechanisms of neuronal demise. IEG = Immediate early genes
Despite increased sensitivity of specific neuronal populations
JNK = c-Jun N-terminal kinase
due to a pro-apoptotic environment in the brain, the death
cascade introduced by many noxious factors may be coun- LOAD = Late onset AD
teracted by other cellular mechanisms which limit the acti-
MASK = Mitogen-activated protein kinase
vation of various apoptotic triggers, suppress oxidoradicals
and stabilize calcium homeostasis and mitochondrial func- MPT = Mitochondrial permeability transition pore
tion. Extrapolating the welth of data gained from typical NFT = Neurofibrillary tangle
cultured to in vivo neurons has been difficult. This is in part
because post-mitotic neurons have different signal transduc- NIAP = Neuronal inhibitor of apoptosis protein
tion systems active than do immature ones and because neu- PAR-4 = Prostate apoptosis response-4
rons embedded in their tissue environment can really power-
ful protective mechanisms to counter disease insults, mecha- PARP-1 = poly(ADP-ribose) polymerase-1
nisms that are hardly available to the cultured cell. The limi- PCD = Programed cell death
tations of putative triggering and promoting factors involved
in neurodegeneration might present disease progression. PHF = Paired helical filament
Postmortem analysis can bridge some but not all of these PP2A = Phosphatae 2A
knowledge gaps, but the results are still controversial, and
the factors and mechanisms that drive this yin-yan balance of PS = Presenilin
neuronel survival and death need to be further elucidated. PTP = Permeability tansition pore
SMAC = Second mitochondrial derived activator of
LIST OF ABBREVIATIONS
caspases
AD = Alzheimer disease tg = Trangenic
AIF = Apoptosis inducing factor TNF = Tumor necrosis factor
APAF1 = Apoptosis protease-inducing factor 1 TNFR = Tumor necrosis factor receptor
APOE = Apolipoprotein E TRADD = TNFR associated death domain protein
APP = Amyloid precursor protein XIAP = X-linked IAP
Aβ = β Amyloid peptide
REFERENCES
ATP = Adenosine triphosphate
[1] Ashe PC and Berry MD. Apoptotic signaling cascades. Prog Neu-
BAR = Bifunctional apoptosis regulator ropsychopharmacol Biol Psychiatry 27: 199-214 (2003).
Challenges in Neuronal Apoptosis Current Alzheimer Research, 2006, Vol. 3, No. 4 387
[2] Becker EB and Bonni A. Cell cycle regulation of neuronal apopto- sis in neurodegenerative diseases: from basic research to clinical
sis in development and disease. Prog Neurobiol 72: 1-25 (2004). application. Neurodeg Dis 1: 9-19 (2004).
[3] Kerr JF, Wyllie AH and Currie AR. Apoptosis: a basic biological [31] Newman J, Rissman RA, Sarsoza F, Kim RC, Dick M, Bennett
phenomenon with wide-ranging implications in tissue kinetics. Br J DA, et al. Csp-cleaved tau accumulation in neurodegenerative dis-
Cancer 26: 239-257 (1972). eases associated with tau and alpha-synuclein pathology. Acta Neu-
[4] Kuhtreiber WM, Hayashi T, Dale EA and Faustman DL. Central ropathol (Berl) 110: 135-144 (2005).
role of defective apoptosis in autoimmunity. J Mol Endocrinol 31: [32] Rissman RA, Poon WW, Blurton-Jones M, Oddo S, Torp R, Vitek
373-399 (2003). MP, et al. Csp-cleavage of tau is an early event in Alzheimer dis-
[5] Lorenz HM, Herrmann M, Winkler T, Gaipl U and Kalden JR. ease tangle pathology. J Clin Invest 114: 121-130 (2004).
Role of apoptosis in autoimmunity. Apoptosis 5: 443-449 (2000). [33] Roth KA. Caspases, apoptosis, and Alzheimer disease: causation,
[6] Flemming W. Ueber die Bildung von Richtungsfiguren in correlation, and confusion. J Neuropathol Exp Neurol 60: 829-838
Saeugethiereiern beim Untergang Graaf'scher Follikel. Arch Anat (2001).
Physiol: 221-224 (1885). [34] Sathasivam S, Ince PG and Shaw PJ. Apoptosis in amyotrophic
[7] Clarke PGH. Apoptosis versus necrosis. In 'Cell death and disease lateral sclerosis: a review of the evidence. Neuropathol Appl Neu-
of the nervous system' (Eds Koliatsosue M, Ratan RR). Humana robiol 27: 257-274 (2001).
Press, Totowa, NY, pp. 3-28 (1999). [35] Schulz JB and Falkenburger BH. Neuronal pathology in Parkin-
[8] Levin S, Bucci TJ, Cohen SM, Fix AS, Hardisty JF, LeGrand EK, son's disease. Cell Tissue Res 318: 135-147 (2004).
et al. The nomenclature of cell death: recommendations of an ad [36] Tatton WG and Olanow CW. Apoptosis in neurodegenerative
hoc Committee of the Society of Toxicologic Pathologists. Toxicol diseases: the role of mitochondria. Biochim Biophys Acta 1410:
Pathol 27: 484-490 (1999). 195-213 (1999).
[9] Wyllie AH, Kerr JF and Currie AR. Cell death: the significance of [37] Vis JC, Schipper E, de Boer-van Huizen RT, Verbeek MM, de
apoptosis. Int Rev Cytol 68: 251-306 (1980). Waal RM, Wesseling P, et al. Expression pattern of apoptosis-
[10] Almeida CJ and Linden R. Phagocytosis of apoptotic cells: a matter related markers in Huntington's disease. Acta Neuropathol (Berl)
of balance. Cell Mol Life Sci: (2005). 109: 321-328 (2005).
[11] Klionsky DJ and Emr SD. Autophagy as a regulated pathway of [38] Cotman CW. Apoptosis decision cascades and neuronal degenera-
cellular degradation. Science 290: 1717-1721 (2000). tion in Alzheimer's disease. Neurobiol Aging 19: S29-32 (1998).
[12] Nixon RA, Yu WH, Cuervo AM, Kumar A, Yang DS and Peterhoff [39] Cotman CW, Poon WW, Rissman RA and Blurton-Jones M. The
C. Autophagy in Alzheimer's disease: failure of a neuroprotective role of Csp cleavage of tau in Alzheimer disease neuropathology. J
mechanism. In 'Recent Advances in Alzheimer's Disease and Re- Neuropathol Exp Neurol 64: 104-112 (2005).
lated Disorders 2004' (Eds Iqbal K, Winblad B). Alzheimer's Asso- [40] Cotman CW, Qian HY and Anderson AJ. Cellular signaling path-
ciation, Chicago, IL, pp. 190-201 (2005). ways in neuronal apoptosis. Role in neurodegeneration and Alz-
[13] Gorman AM, Ceccatelli S and Orrenius S. Role of mitochondria in heimer's disease. In 'Cerebral Signal Transduction. From First to
neuronal apoptosis. Dev Neurosci 22: 348-358 (2000). Fourth Messengers' (Ed Reith MEA). Humana Press, Totowa, NJ,
[14] Nicotera P, Leist M and Ferrando-May E. Apoptosis and necrosis: pp. 175-206 (2000).
different execution of the same death. Biochem Soc Symp 66: 69- [41] Jellinger KA. General aspects of neurodegeneration. J Neural
73 (1999). Transm Suppl: 101-144 (2003).
[15] Desagher S and Martinou JC. Mitochondria as the central control [42] Hirsch EC, Hunot S, Faucheux B, Agid Y, Mizuno Y, Mochizuki
point of apoptosis. Trends Cell Biol 10: 369-377 (2000). H, et al. Dopaminergic neurons degenerate by apoptosis in Parkin-
[16] Jellinger KA and Stadelmann C. Problems of cell death in neurode- son's disease. Mov Disord 14: 383-385 (1999).
generation and Alzheimer's Disease. J Alzheimers Dis 3: 31-40 [43] Mattson MP. Mechanisms of neuronal apoptosis and excitotoxicity.
(2001). In 'Pathogenesis of Neurodegenerative Disorders' (Ed Mattson
[17] Portera-Cailliau C, Price DL and Martin LJ. Excitotoxic neuronal MP). Humana Press, Totowa, NJ, pp. 1-20 (2001).
death in the immature brain is an apoptosis-necrosis morphological [44] Raina AK, Hochman A, Zhu X, Rottkamp CA, Nunomura A,
continuum. J Comp Neurol 378: 70-87 (1997). Siedlak SL, et al. Abortive apoptosis in Alzheimer's disease. Acta
[18] Martin LJ. Neuronal cell death in nervous system development, Neuropathol (Berl) 101: 305-310 (2001).
disease, and injury (Review). Int J Mol Med 7: 455-478 (2001). [45] Reed JC. Mechanisms of apoptosis. Am J Pathol 157: 1415-1430
[19] Jellinger KA and Stadelmann CH. The enigma of cell death in (2000).
neurodegenerative disorders. J Neural Transm Suppl: 21-36 (2000). [46] Prunell GF and Troy CM. Balancing neuronal death. J Neurosci
[20] Graeber MB and Moran LB. Mechanisms of cell death in neurode- Res 78: 1-6 (2004).
generative diseases: fashion, fiction, and facts. Brain Pathol 12: [47] Agerman K, Baudet C, Fundin B, Willson C and Ernfors P. At-
385-390 (2002). tenuation of a Csp-3 dependent cell death in NT4- and p75-
[21] Lockshin RA and Zakeri Z. Apoptosis, autophagy, and more. Int J deficient embryonic sensory neurons. Mol Cell Neurosci 16: 258-
Biochem Cell Biol 36: 2405-2419 (2004). 268 (2000).
[22] Nicotera P and Melino G. Regulation of the apoptosis-necrosis [48] Ashkenazi A and Dixit VM. Death receptors: signaling and modu-
switch. Oncogene 2757-65: 2757-2765 (2004). lation. Science 281: 1305-1308 (1998).
[23] Yuan J, Lipinski M and Degterev A. Diversity in the mechanisms [49] LeBlanc AC, ed Apoptosis Techniques and Protocols, 2nd Edition.
of neuronal cell death. Neuron 40: 401-413 (2003). Totowa, NJ, Humana Press (2002).
[23a] Harwood SM, Yaqoob MM and Allen DA. Caspase and calpain [50] Reed JC. Apoptosis-based therapies. Nat Rev Drug Discov 1: 111-
function in cell death: bridging the gap between apoptosis and 121 (2002).
necrosis. Ann Clin Biochem 42: 415-431 (2005). [51] Bratton SB and Cohen GM. Apoptotic death sensor: an organelle's
[24] Jellinger KA. Apoptosis vs nonapoptotic mechanisms in neurode- alter ego? Trends Pharmacol Sci 22: 306-315 (2001).
generation. In 'Neuroinflammation. 2nd Edition' (Ed Wood PL). [52] Gerhardt E, Kugler S, Leist M, Beier C, Berliocchi L, Volbracht C,
Human Press, Totowa, NJ, pp 29-88 (2003). et al. Cascade of Csp activation in potassium-deprived cerebellar
[25] Bredesen DE. Apoptosis: overview and signal transduction path- granule neurons: targets for treatment with peptide and protein in-
ways. J Neurotrauma 17: 801-810 (2000). hibitors of apoptosis. Mol Cell Neurosci 17: 717-731 (2001).
[26] Wyllie AH. Apoptosis: an overview. Br Med Bull 53: 451-465 [52a] Guo H, Petrin D, Zhang Y, Bergeron C, Goodyer CG and LeBlanc
(1997). AC. Caspase-1 activation of caspase-6 in human apoptotic neurons.
[27] Yuan J and Yankner BA. Apoptosis in the nervous system. Nature Cell Death Differ 13: 285-292 (2006).
407: 802-809 (2000). [53] Velier JJ, Ellison JA, Kikly KK, Spera PA, Barone FC and
[28] Ayala-Grosso C, Ng G, Roy S and Robertson GS. Csp-cleaved Feuerstein GZ. Csp-8 and Csp-3 are expressed by different popula-
amyloid precursor protein in Alzheimer's disease. Brain Pathol 12: tions of cortical neurons undergoing delayed cell death after focal
430-441 (2002). stroke in the rat. J Neurosci 19: 5932-5941 (1999).
[29] Faherty CJ and Smeyne RJ. Cell death in Parkinson's disease. In [54] Budd SL, Tenneti L, Lishnak T and Lipton SA. Mitochondrial and
'Parkinson's Disease' (Eds Ebadi M, Pfeiffer RF). CRC Press, Boca extramitochondrial apoptotic signaling pathways in cerebrocortical
Raton, London, New York, Washington, pp 523-535 (2005). neurons. Proc Natl Acad Sci U S A 97: 6161-6166 (2000).
[30] Kermer P, Liman J, Weishaupt JH and Baehr M. Neuronal apopto- [54a] Hetz CA, Torres V and Quest AF. Beyond apoptosis: nonapoptotic
388 Current Alzheimer Research, 2006, Vol. 3, No. 4 Kurt A. Jellinger
cell death in physiology and disease. Biochem Cell Biol 83: 579- [77] Ramadan S, Terrinoni A, Catani MV, Sayan AE, Knight RA,
588 (2005). Mueller M, et al. p73 induces apoptosis by different mechanisms.
[55] Kroemer G and Martin SJ. Csp-independent cell death. Nat Med Biochem Biophys Res Commun 331: 713-717 (2005).
11: 725-730 (2005). [78] Lin L, Ye Y and Zakeri Z. p53, Apaf-1, Csp-3, and -9 are dispensa-
[55a] Armstrong JS. Mitochondrial membrane permeabilization: the sine ble for Cdk5 activation during cell death. Cell Death Differ: 13:
qua non for cell death. Bioessays 28: 253-260 (2006). 141-150 (2005).
[56] Krieger C and Duchen MR. Mitochondria, Ca2+ and neurodegen- [78a] Krantic S, Mechawar N, Reix S and Quirion R. Molecular basis of
erative disease. Eur J Pharmacol 447: 177-188 (2002). programmed cell death involved in neurodegeneration. Trends
[57] Jordan J, Cena V and Prehn JH. Mitochondrial control of neuron Neurosci 28: 670-676 (2005).
death and its role in neurodegenerative disorders. J Physiol Bio- [79] Davydov V, Hansen LA and Shackelford DA. Is DNA repair com-
chem 59: 129-141 (2003). promised in Alzheimer's disease? Neurobiol Aging 24: 953-968
[58] Gross A, McDonnell JM and Korsmeyer SJ. BCL-2 family mem- (2003).
bers and the mitochondria in apoptosis. Genes Dev 13: 1899-1911 [80] Culmsee C, Bondada S and Mattson MP. Hippocampal neurons of
(1999). mice deficient in DNA-dependent protein kinase exhibit increased
[59] Kroemer G and Reed JC. Mitochondrial control of cell death. Nat vulnerability to DNA damage, oxidative stress and excitotoxicity.
Med 6: 513-519 (2000). Brain Res Mol Brain Res 87: 257-262 (2001).
[60] Zou H, Li Y, Liu X and Wang X. An APAF-1.cytochrome c mul- [81] Gilmore EC, Nowakowski RS, Caviness VS, Jr. and Herrup K. Cell
timeric complex is a functional apoptosome that activates pro- birth, cell death, cell diversity and DNA breaks: how do they all fit
caspase-9. J Biol Chem 274: 11549-11556 (1999). together? Trends Neurosci 23: 100-105 (2000).
[61] Qin H, Srinivasula SM, Wu G, Fernandes-Alnemri T, Alnemri ES [82] Du C, Fang M, Li Y, Li L and Wang X. Smac, a mitochondrial
and Shi Y. Structural basis of procaspase-9 recruitment by the protein that promotes cytochrome c-dependent Csp activation by
apoptotic protease-activating factor 1. Nature 399: 549-557 (1999). eliminating IAP inhibition. Cell 102: 33-42 (2000).
[62] Salvesen GS and Dixit VM. Csp activation: the induced-proximity [83] Jaattela M, Wissing D, Kokholm K, Kallunki T and Egeblad M.
model. Proc Natl Acad Sci U S A 96: 10964-10967 (1999). Hsp70 exerts its anti-apoptotic function downstream of Csp-3-like
[63] Hengartner MO. The biochemistry of apoptosis. Nature 407: 770- proteases. Embo J 17: 6124-6134 (1998).
776 (2000). [84] Volbracht C, Leist M, Kolb SA and Nicotera P. Apoptosis in Csp-
[64] Stefanis L. Csp-dependent and -independent neuronal death: two inhibited neurons. Mol Med 7: 36-48 (2001).
distinct pathways to neuronal injury. Neuroscientist 11: 50-62 [85] Ha HC and Snyder SH. Poly(ADP-ribose) polymerase-1 in the
(2005). nervous system. Neurobiol Dis 7: 225-239 (2000).
[64a] Kim R, Emi M, Tanabe K, Murakami S, Uchida Y and Arihiro K. [86] Sheline CT, Wang H, Cai AL, Dawson VL and Choi DW. In-
Regulation and interplay of apoptotic and non-apoptotic cell death. volvement of poly ADP ribosyl polymerase-1 in acute but not
J Pathol 208: 319-326 (2006). chronic zinc toxicity. Eur J Neurosci 18: 1402-1409 (2003).
[65] Joza N, Susin SA, Daugas E, Stanford WL, Cho SK, Li CY, et al. [87] Herdegen T, Skene P and Bahr M. The c-Jun transcription factor--
Essential role of the mitochondrial apoptosis-inducing factor in bipotential mediator of neuronal death, survival and regeneration.
programmed cell death. Nature 410: 549-554 (2001). Trends Neurosci 20: 227-231 (1997).
[66] Longuet M, Serduc R and Riva C. Implication of bax in apoptosis [88] Pozas E, Ballabriga J, Planas AM and Ferrer I. Kainic acid-induced
depends on microtubule network mobility. Int J Oncol 25: 309-317 excitotoxicity is associated with a complex c-Fos and c-Jun re-
(2004). sponse which does not preclude either cell death or survival. J Neu-
[67] Roth KA, Kuan C, Haydar TF, D'Sa-Eipper C, Shindler KS, Zheng robiol 33: 232-246 (1997).
TS, et al. Epistatic and independent functions of Csp-3 and Bcl- [89] Bozyczko-Coyne D, Saporito MS and Hudkins RL. Targeting the
X(L) in developmental programmed cell death. Proc Natl Acad Sci JNK pathway for therapeutic benefit in CNS disease. Curr Drug
U S A 97: 466-471 (2000). Targets CNS Neurol Disord 1: 31-49 (2002).
[68] Cregan SP, MacLaurin JG, Craig CG, Robertson GS, Nicholson [90] Mattson MP. Apoptosis in neurodegenerative disorders. Nat Rev
DW, Park DS, et al. Bax-dependent Csp-3 activation is a key de- Mol Cell Biol 1: 120-129 (2000).
terminant in p53-induced apoptosis in neurons. J Neurosci 19: [90a] Langley B and Ratan RR. Oxidative stress-induced death in the
7860-7869 (1999). nervous system: cell cycle dependent or independent? J Neurosci
[69] Finucane DM, Bossy-Wetzel E, Waterhouse NJ, Cotter TG and Res 77: 621-629 (2004).
Green DR. Bax-induced Csp activation and apoptosis via cyto- [90b] Kruman, II, Wersto RP, Cardozo-Pelaez F, Smilenov L, Chan SL,
chrome c release from mitochondria is inhibitable by Bcl-xL. J Biol Chrest FJ, et al. Cell cycle activation linked to neuronal cell death
Chem 274: 2225-2233 (1999). initiated by DNA damage. Neuron 41: 549-561 (2004).
[70] Takayama S and Reed JC. Molecular chaperone targeting and [91] Wolozin B and Behl C. Mechanisms of neurodegenerative disor-
regulation by BAG family proteins. Nat Cell Biol 3: E237-241 ders: Part 1: protein aggregates. Arch Neurol 57: 793-796 (2000).
(2001). [92] Jellinger KA. Cell death mechanisms in neurodegeneration. J Cell
[71] Kermer P, Digicaylioglu MH, Kaul M, Zapata JM, Krajewska M, Mol Med 5: 1-17 (2001).
Stenner-Liewen F, et al. BAG1 over-expression in brain protects [93] Duda JE, Lee VM and Trojanowski JQ. Neuropathology of synu-
against stroke. Brain Pathol 13: 495-506 (2003). clein aggregates. J Neurosci Res 61: 121-127 (2000).
[72] Roth W, Kermer P, Krajewska M, Welsh K, Davis S, Krajewski S, [94] Kahns S, Lykkebo S, Jakobsen LD, Nielsen MS and Jensen PH.
et al. Bifunctional apoptosis inhibitor (BAR) protects neurons from Csp-mediated parkin cleavage in apoptotic cell death. J Biol Chem
diverse cell death pathways. Cell Death Differ 10: 1178-1187 277: 15303-15308 (2002).
(2003). [95] Lee SS, Kim YM, Junn E, Lee G, Park KH, Tanaka M, et al. Cell
[73] Boonman Z and Isacson O. Apoptosis in neuronal development and cycle aberrations by alpha-synuclein over-expression and cyclin B
transplantation: role of caspases and trophic factors. Exp Neurol immunoreactivity in Lewy bodies. Neurobiol Aging 24: 687-696
156: 1-15 (1999). (2003).
[73a] Butts BD, Hudson HR, Linseman DA, Le SS, Ryan KR, Bouchard [96] Lucking CB and Brice A. Alpha-synuclein and Parkinson's disease.
RJ, et al. Proteasome inhibition elicits a biphasic effect on neuronal Cell Mol Life Sci 57: 1894-1908 (2000).
apoptosis via differential regulation of pro-survival and pro- [96a] Kim R, Emi M, Tanabe K and Murakami S. Role of the unfolded
apoptotic transcription factors. Mol Cell Neurosci 30: 279-289 protein response in cell death. Apoptosis 11: 5-13 (2006).
(2005). [96b] Fadeel B and Orrenius S. Apoptosis: a basic biological
[74] Mattson MP, Duan W, Chan SL and Camandola S. Par-4: an phenomenon with wide-ranging implications in human disease. J
emerging pivotal player in neuronal apoptosis and neurodegenera- Intern Med 258: 479-517 (2005).
tive disorders. J Mol Neurosci 13: 17-30 (1999). [97] Sathasivam S and Shaw PJ. Apoptosis in amyotrophic lateral scle-
[75] Culmsee C and Mattson MP. p53 in neuronal apoptosis. Biochem rosis--what is the evidence? Lancet Neurol 4: 500-509 (2005).
Biophys Res Commun 331: 761-777 (2005). [98] Knopman DS. Alzheimer type dementia. In 'Neurodegeneration.
[76] Martin LJ, Chen K and Liu Z. Adult motor neuron apoptosis is The Molecular Pathology of Dementia and Movement Disorders'
mediated by nitric oxide and Fas death receptor linked by DNA (Ed Dickson DW). ISN Neuropath Press, Basel, pp. 24-39 (2003).
damage and p53 activation. J Neurosci 25: 6449-6459 (2005). [99] Mirra SS and Hyman BT. Aging and dementia. In 'Greenfield's
Challenges in Neuronal Apoptosis Current Alzheimer Research, 2006, Vol. 3, No. 4 389
Neuropathology, 7th Ed.' (Eds Graham DI, Lantos PL). E. Arnold, brillary tangles in P301l tau transgenic mice induced by Abeta 42
London, pp. 195-271 (2002). fibrils. Science 293: 1491-1495 (2001).
[100] Braak H and Braak E. Neuropathological stageing of Alzheimer- [123] Oddo S, Caccamo A, Kitazawa M, Tseng BP and LaFerla FM.
related changes. Acta Neuropathol (Berl) 82: 239-259. (1991). Amyloid deposition precedes tangle formation in a triple transgenic
[101] McKeel DW, Price JL, Miller JP, Grant EA, Xiong C, Berg L, et al. model of Alzheimer's disease. Neurobiol Aging 24: 1063-1070
Neuropathologic criteria for diagnosing Alzheimer disease in per- (2003).
sons with pure dementia of Alzheimer type. J Neuropathol Exp [124] Ferrer I, Gomez-Isla T, Puig B, Freixes M, Ribé E, Dalfó E, et al.
Neurol 63: 1028-1037 (2004). Current advances on different kinases involved in tau phosphoryla-
[102] Duyckaerts C and Dickson DW. Neuropathology of Alzheimer´s tion, and implications in Alzheimer’s disease and tauopathies. Curr
disease. In 'Neurodegeneration. The Molecular Pathology of De- Alzheimer Res 2: 3-18 (2005).
mentia and Movement Disorders' (Ed Dickson DW). ISN Neuro- [125] Liu T, Perry G, Chan HW, Verdile G, Martins RN, Smith MA, et
path Press Basel, pp. 47-65 (2003). al. Amyloid-beta-induced toxicity of primary neurons is dependent
[103] Morris JH and Nagy Z. Alzheimer's disease. In 'The Neuropathol- upon differentiation-associated increases in tau and cyclin-
ogy of Dementia' (Eds Esiri M, Lee VMY, Trojanowski JQ). Cam- dependent kinase 5 expression. J Neurochem 88: 554-563 (2004).
bridge University Press, Cambridge, pp. 161-206 (2004). [126] Kins S, Kurosinski P, Nitsch RM and Gotz J. Activation of the
[104] Masters CL and Beyreuther K. Molecular Pathogenesis of Alz- ERK and JNK signaling pathways caused by neuron-specific inhi-
heimer´s disease. In 'Neurodegeneration. The Molecular Pathology bition of PP2A in transgenic mice. Am J Pathol 163: 833-843
of Dementia and Movement Disorders' (Ed Dickson DW). ISN (2003).
Neuropath Press Basel, pp. 69-73 (2003). [127] Xie L, Helmerhorst E, Taddei K, Plewright B, Van Bronswijk W
[105] Buée L, Bussiere T, Buee-Scherrer V, Delacourte A and Hof PR. and Martins R. Alzheimer's beta-amyloid peptides compete for in-
Tau protein isoforms, phosphorylation and role in neurodegenera- sulin binding to the insulin receptor. J Neurosci 22: RC221 (2002).
tive disorders. Brain Res Brain Res Rev 33: 95-130 (2000). [128] Ferrer I, Blanco R, Carmona M and Puig B. Phosphorylated mito-
[106] Delacourte A, Sergeant N, Champain D, Wattez A, Maurage CA, gen-activated protein kinase (MAPK/ERK-P), protein kinase of 38
Lebert F, et al. Nonoverlapping but synergetic tau and APP pa- kDa (p38-P), stress-activated protein kinase (SAPK/JNK-P), and
thologies in sporadic Alzheimer's disease. Neurology 59: 398-407. calcium/calmodulin-dependent kinase II (CaM kinase II) are differ-
(2002). entially expressed in tau deposits in neurons and glial cells in
[107] Hardy J and Selkoe DJ. The amyloid hypothesis of Alzheimer's tauopathies. J Neural Transm 108: 1397-1415 (2001).
disease: progress and problems on the road to therapeutics. Science [129] Ferrer I, Blanco R, Carmona M, Ribera R, Goutan E, Puig B, et al.
297: 353-356 (2002). Phosphorylated map kinase (ERK1, ERK2) expression is associ-
[108] LeVine H, 3rd. The Amyloid Hypothesis and the clearance and ated with early tau deposition in neurones and glial cells, but not
degradation of Alzheimer's beta-peptide. J Alzheimers Dis 6: 303- with increased nuclear DNA vulnerability and cell death, in Alz-
314 (2004). heimer disease, Pick's disease, progressive supranuclear palsy and
[109] Ferrari A, Hoerndli F, Baechi T, Nitsch RM and Gotz J. beta- corticobasal degeneration. Brain Pathol 11: 144-158 (2001).
Amyloid induces paired helical filament-like tau filaments in tissue [130] Fasulo L, Ugolini G, Visintin M, Bradbury A, Brancolini C,
culture. J Biol Chem 278: 40162-40168 (2003). Verzillo V, et al. The neuronal microtubule-associated protein tau
[110] Selznick LA, Zheng TS, Flavell RA, Rakic P and Roth KA. Amy- is a substrate for Csp-3 and an effector of apoptosis. J Neurochem
loid beta-induced neuronal death is bax-dependent but Csp- 75: 624-633 (2000).
independent. J Neuropathol Exp Neurol 59: 271-279 (2000). [131] Horowitz PM, Patterson KR, Guillozet-Bongaarts AL, Reynolds
[111] Allen JW, Eldadah BA, Huang X, Knoblach SM and Faden AI. MR, Carroll CA, Weintraub ST, et al. Early N-terminal changes
Multiple caspases are involved in beta-amyloid-induced neuronal and Csp-6 cleavage of tau in Alzheimer's disease. J Neurosci 24:
apoptosis. J Neurosci Res 65: 45-53 (2001). 7895-7902 (2004).
[112] Pereira C, Ferreiro E, Cardoso SM and de Oliveira CR. Cell degen- [132] Kobayashi K, Nakano H, Hayashi M, Shimazaki M, Fukutani Y,
eration induced by amyloid-beta peptides: implications for Alz- Sasaki K, et al. Association of phosphorylation site of tau protein
heimer's disease. J Mol Neurosci 23: 97-104 (2004). with neuronal apoptosis in Alzheimer's disease. J Neurol Sci 208:
[113] Cardoso SM and Oliveira CR. The role of calcineurin in amyloid- 17-24 (2003).
beta-peptides-mediated cell death. Brain Res 1050: 1-7 (2005). [133] Kang HJ, Yoon WJ, Moon GJ, Kim DY, Sohn S, Kwon HJ, et al.
[114] Gervais FG, Xu D, Robertson GS, Vaillancourt JP, Zhu Y, Huang Csp-3-mediated cleavage of PHF-1 tau during apoptosis irrespec-
J, et al. Involvement of caspases in proteolytic cleavage of Alz- tive of excitotoxicity and oxidative stress: an implication to Alz-
heimer's amyloid-beta precursor protein and amyloidogenic A beta heimer's disease. Neurobiol Dis 18: 450-458 (2005).
peptide formation. Cell 97: 395-406 (1999). [134] Masliah E, Mallory M, Alford M, Tanaka S and Hansen LA. Csp
[115] Rohn TT, Rissman RA, Davis MC, Kim YE, Cotman CW and dependent DNA fragmentation might be associated with excito-
Head E. Csp-9 activation and Csp cleavage of tau in the Alz- toxicity in Alzheimer disease. J Neuropathol Exp Neurol 57: 1041-
heimer's disease brain. Neurobiol Dis 11: 341-354 (2002). 1052 (1998).
[116] Rohn TT, Rissman RA, Head E and Cotman CW. Csp Activation [135] Gastard MC, Troncoso JC and Koliatsos VE. Csp activation in the
in the Alzheimer's Disease Brain: Tortuous and Torturous. Drug limbic cortex of subjects with early Alzheimer's disease. Ann Neu-
News Perspect 15: 549-557 (2002). rol 54: 393-398 (2003).
[117] Fuentealba RA, Farias G, Scheu J, Bronfman M, Marzolo MP and [136] Pompl PN, Yemul S, Xiang Z, Ho L, Haroutunian V, Purohit D, et
Inestrosa NC. Signal transduction during amyloid-beta-peptide al. Csp gene expression in the brain as a function of the clinical
neurotoxicity: role in Alzheimer disease. Brain Res Brain Res Rev progression of Alzheimer disease. Arch Neurol 60: 369-376 (2003).
47: 275-289 (2004). [137] Cribbs DH, Poon WW, Rissman RA and Blurton-Jones M. Csp-
[118] Gamblin TC, Chen F, Zambrano A, Abraha A, Lagalwar S, Guil- mediated degeneration in Alzheimer's disease. Am J Pathol 165:
lozet AL, et al. Csp cleavage of tau: linking amyloid and neurofi- 353-355 (2004).
brillary tangles in Alzheimer's disease. Proc Natl Acad Sci U S A [138] Rametti A, Esclaire F, Yardin C and Terro F. Linking alterations in
100: 10032-10037 (2003). tau phosphorylation and cleavage during neuronal apoptosis. J Biol
[119] Guo H, Albrecht S, Bourdeau M, Petzke T, Bergeron C and Le- Chem 279: 54518-54528 (2004).
Blanc AC. Active Csp-6 and Csp-6-cleaved tau in neuropil threads, [139] Dickson DW. Apoptotic mechanisms in Alzheimer neurofibrillary
neuritic plaques, and neurofibrillary tangles of Alzheimer's disease. degeneration: cause or effect? J Clin Invest 114: 23-27 (2004).
Am J Pathol 165: 523-531 (2004). [140] Sun XM, Butterworth M, MacFarlane M, Dubiel W, Ciechanover
[120] Oddo S, Billings L, Kesslak JP, Cribbs DH and LaFerla FM. Abeta A and Cohen GM. Csp activation inhibits proteasome function
immunotherapy leads to clearance of early, but not late, hyper- during apoptosis. Mol Cell 14: 81-93 (2004).
phosphorylated tau aggregates via the proteasome. Neuron 43: 321- [141] Braak H and Braak E. Frequency of stages of Alzheimer-related
332 (2004). lesions in different age categories. Neurobiol Aging 18: 351-357
[121] Chung CW, Song YH, Kim IK, Yoon WJ, Ryu BR, Jo DG, et al. (1997).
Proapoptotic effects of tau cleavage product generated by Csp-3. [142] Duyckaerts C and Hauw JJ. Prevalence, incidence and duration of
Neurobiol Dis 8: 162-172 (2001). Braak's stages in the general population: can we know? Neurobiol
[122] Gotz J, Chen F, van Dorpe J and Nitsch RM. Formation of neurofi- Aging 18: 362-369; discussion 389-392 (1997).
390 Current Alzheimer Research, 2006, Vol. 3, No. 4 Kurt A. Jellinger
neurodegeneration in Alzheimer's disease. FASEB J 19: 255-257 the relative resistance of parvalbumin and calretinin immunoreac-
(2005). tive neocortical neurons to the pathology of Alzheimer's disease.
[184] de la Monte SM, Luong T, Neely TR, Robinson D and Wands JR. Exp Neurol 145: 295-302 (1997).
Mitochondrial DNA damage as a mechanism of cell loss in Alz- [188] Vickers JC, Tan A and Dickson TC. Direct determination of the
heimer's disease. Lab Invest 80: 1323-1335 (2000). proportion of intra- and extra-cellular neocortical neurofibrillary
[185] Perry G, Nunomura A, Lucassen P, Lassmann H and Smith MA. tangles in Alzheimer's disease. Brain Res 971: 135-137 (2003).
Apoptosis and Alzheimer's disease. Science 282: 1268-1269 [189] Andorfer C, Acker CM, Kress Y, Hof PR, Duff K and Davies P.
(1998). Cell-cycle reentry and cell death in transgenic mice expressing
[186] Morsch R, Simon W and Coleman PD. Neurons may live for dec- nonmutant human tau isoforms. J Neurosci 25: 5446-5454 (2005).
ades with neurofibrillary tangles. J Neuropathol Exp Neurol 58: [190] Santacruz K, Lewis J, Spires T, Paulson J, Kotilinek L, Ingelsson
188-197 (1999). M, et al. Tau suppression in a neurodegenerative mouse model im-
[187] Sampson VL, Morrison JH and Vickers JC. The cellular basis for proves memory function. Science 309: 476-481 (2005).