Amyloid precursor protein and mitochondrial dysfunction...

Amyloid precursor protein and mitochondrial dysfunction in Alzheimer's disease

Growing evidence suggests that mitochondrial dysfunction is one of the key intracellular lesions associated with the pathogenesis of Alzheimer's disease (AD). Mitochondria, the powerhouses of the cell, participate in a number of physiological functions that include calcium homeostasis, signal transduction, and apoptosis. However, the pathophysiological mechanisms underlying the decline of mitochondrial vital functions leading to the dysfunction of mitochondria during AD are not well understood. Recent literature has observed the accumulation of Alzheimer's amyloid precursor protein (APP) and its C-terminal—cleaved product ß-amyloid (Aß) in the mitochondrial compartment. Furthermore, evidence also implicates that the accumulation of full-length APP and Aß in the mitochondrial compartment has a causative role in impairing mitochondrial physiological functions. Here, we review the mode of mitochondrial transport of full-length APP and Aß and its pathological implications in bringing about mitochondrial dysfunction as seen in AD.

Anandatheerthavarada HK, Devi L. Amyloid precursor protein and mitochondrial dysfunction in Alzheimer's disease. Neuroscientist 2007;13(6): 626-638.

Department of Animal Biology, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, ann1234@vet.upenn.edu


FULL PAPER

The most striking clinicopathological features of
Alzheimer’s disease (AD) are the extracellular deposi-
tion of amyloid plaques,intracellular neurofibrillary tan-
gles, progressive loss of neurons, and dementia (Price
and Sisodia 1998; Selkoe 1999; Thinakaran 1999).
Amyloid plaques are composed of deposits of β-amyloid
(Aβ). Aβ is an amino-acid peptide derived from the pro-
teolytic processing of a larger plasma-membrane–bound
protein known as amyloid precursor protein (APP; Price
and Sisodia 1998; Thinakaran 1999; Nunan and Small
2000). A growing number of reports suggests that
besides its well-characterized neuropathological symp-
toms, AD is also thought to be associated with many
intracellular lesions such as perturbation of Ca2+ home-
ostasis, accumulation of Aβ in the secretory pathway,
apoptosis, and mitochondrial dysfunction (Manfredi and
Beal 2000; Mattson 1997; Tabira and others 2002;
Gouras and others 2005; Beal 1998; Sims 1996; Offen
and others 2000; Reddy 2006). However, it is not clear,
to date, whether these intracellular lesions are independ-
ent or interdependent on each other. In addition, AD is
the most common form of dementia and is the fourth
leading cause of death in the developed world. Thus, the
comprehensive understanding of the genesis of lesions
leading to the pathogenesis of AD might be a key issue
in the development of safe pharmacological therapies for
this disease.


AD and Mitochondrial Dysfunction

It is widely known that mitochondria play a crucial role
in many cellular events including energy metabolism,
fatty-acid oxidation, cellular signaling, heme biosyn-
thesis, calcium homeostasis, and apoptosis (Fig. 1).
Furthermore, neurons in particular depend on mitochondr-
ial functions, especially for their energy supply. Hence,
impairment in mitochondrial vital functions may have
serious and deleterious consequences of neuronal physi-
ology. Mitochondrial dysfunction in AD is associated
with increased free-radical generation leading to oxida-
tive stress, decreased cytochrome coxidase activity, and
reduced energy metabolism (Parker 1991; Maurer and
others 2000; Lin and Beal 2006; Reddy and Beal 2005).
It is also suggested that mitochondrial dysfunction may
play a central role in the degeneration and death of
neurons during the pathogenesis of AD (Beal 2005).
Morphological and morphometric analysis of mitochon-
dria by electron microscopy revealed significant changes
in various brain regions such as the hippocampus,frontal
cortex, cerebellum, thalamus, globus pallidus, and locus
coeruleus in AD patients (Baloyannis 2006). The morpho-
logical alterations include a decrease in membrane fluidity
and in the size of mitochondrial cristae. Furthermore,mito-
chondrial alterations in neurons were associated with the
fragmentation of the cisternae of the golgi apparatus
(Baloyannis 2006). However, the factors responsible for
causing intracellular lesions—particularly, decline in
mitochondrial functions—during AD progression are
not clear. Recent studies have suggested that mitochon-
dria may be a direct target of Alzheimer’s disease–asso-
ciated proteins and peptides such as full-length APP,Aβ
peptide, Tau,and ApoE4 (David and others 2005; Chang
and others 2005; Anandatheerthavarada and others 2003;
Devi and others 2006; Keil and others 2004; Lustbader
and others 2004; Caspersen and others 2005; Manczak
and others 2006; Casley and others 2002). The aim of
the present article is to review recent advancements in
the field of AD that suggest an important role for mito-
chondrially targeted APP and its C-terminal proteolytic
product Aβ in disrupting the mitochondrial basic func-
tions including oxidative phosphorylation and protein
import.


________
Fig. 1. Functions of mitochondria. Mitochondria contain two membranes: outer and inner membranes. These two membranes are separated by a space called intermembrane space. The space enclosed by the inner membrane is the matrix compartment that contains soluble proteins. Additionally, the mitochondrial genome, ribosomes, and tRNAs that are needed for protein synthesis are also present in the matrix. Mitochondria participate in the following cellular events: (A) uptake of calcium, which helps maintain the cellular calcium levels; (B) provision of energy and heme moieties in the cell; and (C) initiation of programmed cell death or apoptosis as and when the cell requires.
________



Mitochondrial Functions in Neurons

As far as neuronal physiology is concerned,mitochondria
are considered to be vital organelles for various neuronal
functions, especially at synapses. Mitochondria at
synapses play a major role in energy-demanding neuro-
transmission by generating adenosine triphosphate (ATP)
and maintaining the calcium homeostasis (Nicholls and
Budd 2000; Kann and Kovacs 2006). Hence, synapses
are enriched with mitochondria. In neurons, mitochon-
dria are highly dynamic and mobile. Mitochondria are
transported from the neuronal body to synapses via
microtubule-assisted transport. Therefore, movement and
sequestration of mitochondria at synapses is pivotal for
regulation of synaptic strength. Interestingly, impairment
of synaptic strength is believed to be one of the common-
est lesions in the pathology of various neurodegenerative
diseases including AD. Hence, the events that perturb the
mitochondrial functional machinery and the dynamics
may potentially affect the synaptic physiology. In support
of this notion,a recent study suggested that mitochondrial
alterations were particularly prominent in neurons of
AD patients, which showed the loss of dendritic spines
and contraction of the dendritic arborization (Baloyannis
2006).

Basically,mitochondria emerged from a symbiotic asso-
ciation between a proto-eukaryotic cell and a bacterium.
Mitochondria are double-membranous, self-replicating
organelles with a circular genome of 16.5 kb DNA (Chen
and Chan 2005; Fig. 1). Mitochondria also contain
machinery for transcription, translation, and the protein-
assembly system. Each mitochondrion possesses multiple
copies of mitochondrial DNA, which codes for 13
polypeptides that are part of the mitochondrial electron-
transport chain involved in the oxidative phosphorylation
that generates ATP. Oxidative phosphorylation operates
through five protein complexes embedded in the inner
membrane of the mitochondria. These complexes include
complex I,complex II,complex III,complex IV,and com-
plex V (Fig. 2). The functions of these complexes can be
specifically inhibited by a number of chemicals (Table 1).
Among 13 polypeptides, the mitochondrial genome
encodes seven subunits of complex I, one of complex III,
three of complex IV,and two of complex V (Table 2). The
mitochondrial DNA also encodes the 12S and 16S rRNA
genes and the 22 tRNA genes required for mitochondrial
protein synthesis (Fig. 1). Furthermore, the mitochondrial
genome is more vulnerable to free-radical damage, as
it contains no protective histones. The mitochondrial elec-
tron-transport chain consumes ~85% of the oxygen, and
the disruption of the components of the electron-transport
chain results in increased free-radical production,decreased
ATP production, and the dissipation of membrane reduc-
tion–oxidation potential (Fig. 2). Mitochondria require
a large number of proteins for carrying out their physio-
logical functions, despite the availability of a very small
number of proteins synthesized by their own genome.
Hence, mitochondria depend on proteins coded by the
nuclear genome, which are imported through protein-
import machinery consisting of outer and inner mem-
brane translocases (Fig. 3).

________
Fig. 2. Production of ATP by oxidative phosphorylation. The mitochondrial inner membrane contains five multisubunit protein complexes, which are involved in the electron transport, and generates proton flow that is used to drive ATP synthesis. There are several chemicals, called uncouplers, that cause the dissipation of membrane proton gradients. The electron flow starts from complex I, which accepts electrons from NADH and transfers to ubiquinone. On acceptance of electrons, ubiquinone forms ubiquinol. In addition to this, another pool of ubiquinol comes from complex II, which dehydrogenates succinate to provide electrons to ubiquinone. The two ubiquinol pools generated by complexes I and II diffuse to complex III, which in turn reduces cytochrome c(an intermembrane-space resident heme protein) by pumping electrons from ubiquinol. These electrons are further transferred to complex IV, termed as cytochrome oxidase, which produces 1 H2O per 2 oxidized cytochrome c. This process generates a proton gradient across the membrane. Meanwhile, complex V, known as ATP synthase, converts the proton gradient into ATP from ADP. Complex V requires three protons to form one molecule of ATP. UQ =ubiquinone; Cyto. c =cytochrome c.
________


________
Table 1. Inhibitors of Components of Mitochondrial Oxidative Phosphorylation

Complex I (NADH-dehydrogenase)
- Rotenone
- Ptericidin
- Amytal
- Mercurials
- Demeral

Complex II (Succinate dehydrogenase)
- 3-nitropropionic acid
- Thenoyl trifluroacetone
- Carboxin

Complex III (Ubiquinone cytochrome c oxidoreductase)
- Antimycin

Complex IV (Cytochrome c oxidase)
- Cyanide
- Azide
- Carbon monoxide

Complex V (ATP synthase)
- DCCD
- Oligomycin
___
DCCD=dicyclohexylcarbodiimide.
________


________
Table 2. Proteins Encoded by Mitochondrial Genome

Complex I
- ND1
- ND2
- ND3
- ND4
- ND4L
- ND5
- ND6

Complex II

Complex III
- Cytochrome b

Complex IV
- Subunit I
- Subunit II
- Subunit III

Complex V
- ATPase 6
- ATPase 8
________



Translocation of Proteins to Mitochondria

Mitochondrial protein import is a relatively new but
an emerging field of study in the area of neuroscience
and related neurodegenerative disorders. Recent studies
suggest that either genetic defect in the mitochondrial
import receptors or a physical block in the import chan-
nels may lead to reduced import of some of the mito-
chondrial proteins encoded by the nuclear genome
(Roesch and others 2002; Devi and others 2006). It has
been estimated that approximately 1500 nuclear-encoded
proteins are imported into mammalian mitochondria
under physiological conditions (Taylor and others 2003;
Gabaldon and Huynen 2004). Transport of almost all of
these proteins to mitochondria requires functional outer-
and inner-membrane import channels. In general, mito-
chondrial targeting signals are rich in basic amino acids
and can form amphipathic α-helices. These signals ini-
tially are recognized by outer-membrane import recep-
tors, termed translocase of the outer membrane (TOM).
The TOM machinery consists of major outer-membrane
peripheral-surface receptors such as TOM70, TOM20,
and TOM22 as well as pore-forming TOM40, termed
general-import pore (GIP; Fig. 3). Following the sequen-
tial recognition of these signals by peripheral TOM recep-
tors,proteins are transported through TOM40. Depending
on the destination, proteins are further recognized by one
of the two translocases of the inner membrane (TIM)
complexes—namely, TIM23 and TIM22. Proteins tar-
geted to the matrix are recognized by the TIM23 complex,
which consists of channel-forming TIM23 and peripheral
TIM17. Additionally,TIM23 import machinery is further
assisted by TIM44 and the HSP70 complex in pulling
these proteins into the matrix. On the other hand, poly-
typic inner-membrane proteins are recognized by the
TIM22 import complex that consists of channel-forming
TIM22,TIM18,and TIM54 (Fig. 3). One of the important
features of mitochondrial targeted proteins is that these
proteins should be in an unfolded conformation during the
translocation. Proteins targeted to inner-membrane and
matrix compartments require ATP as well as mitochondr-
ial membrane potential to translocate through mitochon-
drial TOM and TIM channels (Fig. 3).


________
Fig. 3. Description of mitochondrial general import pathway. Mitochondrial protein-import machinery consists of the g. 3. Description of mitochondrial general import pathway. Mitochondrial protein-import machinery consists of the translocase of the outer mitochondrial membrane (TOM) and the translocase of the inner membrane (TIM). These two components of translocation machinery require ATP. The TOM complex mediates translocation of all nuclear-coded mitochondrial proteins across the outer membrane (OM) and is composed of TOM20, TOM22, TOM70, and TOM40. TOM40 is very essential for cell viability and forms a stable complex with one TOM22 and TOM proteins, namely TOM5, TOM6, and TOM7. Next, the two TIM complexes facilitate the translocation of proteins across the inner membrane (IM). The major components of the TIM23 complex are TIM23 and TIM17. TIM23 forms a voltage-activated channel. Importantly, the membrane potential across the IM is essential for the translocation of proteins with N-terminal cleavable signals (I) into the matrix. Interestingly, the TIM23 complex is in direct contact with proteins during their translocation and is assisted by matrix Hsp70 and TIM44, which form an ATP-dependent import motor. The matrix-imported presequence is cleaved by the mitochondrial-processing peptidase (MPP). Proteins with internal targeting signals use the TIM22 complex for the translocation across the IM. Small TIM proteins in the intermembrane space (IMS) help these proteins to cross the aqueous IMS compartment to the IM-embedded TIM22 complex, which is composed of three major membrane proteins, namely TIM18, pore-forming TIM22, and TIM54. Similarly, proteins with chimeric signals (III) also use the TOM40 and TIM23 complexes to cross the OM and IM, respectively.
________



Variations in Mitochondrial Targeting Signals

The majority of mitochondrial proteins are synthesized
on smooth ribosomes with N-terminal extensions enriched
with basic amino acids, which act as targeting signals
(Fig. 3I). These N-terminal signals, particularly those of
proteins targeted to the matrix and inner membrane, are
cleaved by mitochondrial matrix metalloprotease (MMP;
Fig. 3). However, approximately 30% of mitochondrial-
targeted proteins contain internal signals, which are not
cleaved by MMP (Fig. 3II). Such proteins include multiple
membrane-spanning metabolite transporters of the inner
membrane, preproteins directed at the intermembrane
space,and all mitochondrial outer-membrane resident pro-
teins (Endo and others 2003). Interestingly, studies from
our laboratory have identified a novel N-terminal cryptic
mitochondrial targeting sequence that is part of chimeric
signals in a number of xenobiotic metabolizing enzymes
including cytochrome P450 proteins (Fig. 3III). These
chimeric signals contain N-terminal hydrophobic endoplas-
mic reticulum (ER) signals followed by mitochondrial sig-
nals (Fig. 4). By virtue of these signals, P450 proteins are
targeted to both ER and mitochondria. The activation of
cryptic mitochondrial signals is under the control of a num-
ber of physiological factors that vary from protein to pro-
tein. The cryptic mitochondrial targeting signals of
P4501A1 are activated by a cytosolic serine protease,which
cleaves P4501A1 at the 32nd amino acid from the N-termi-
nus to activate the signals (Addya and others 1997).
Alternatively, mitochondrial targeting of P4502B1
(Anandatheerthavarada and others 1999) and 2E1
(Robin and others 2002) are highly regulated by the levels
of PKA-mediated phosphorylation at Ser 129 and Ser 128,
respectively. Although these proteins are targeted to inner
mitochondrial compartments, their mitochondrial signals
are not cleaved by matrix metalloproteases. Importantly,
these proteins faithfully follow the mitochondrial import
paradigms. This type of chimeric signal is also found in APP
(Fig. 4), but the underlying mechanisms involved in the
activation of mitochondrial signals of APP are not known.


________
Fig. 4. Similarities of amyloid precursor protein (APP) N-terminal sequence with chimeric signals. Acomparison of the N-terminal targeting sequence of APP with well-characterized dual targeting signals of cytochrome P4501A1, 2B1, and 2E1 reveals the structural homology among these molecules. The amino terminal of 35 amino acids of APP constitutes a hydrophobic endoplasmic-reticulum (ER) targeting signal followed by a mitochondrial targeting signal spanning between 35 and 67 amino acids. In addition,N-terminal positively charged residues at 40, 44, and 51 of APP are important components of the mitochondrial targeting signal.
________



APP Isoforms and Aβ Processing in Secretory Pathway

It is well known that APP plays a key role in the patho-
genesis of AD. In an attempt to address whether mito-
chondria are directly targeted to APP, our laboratory
observed N-terminal dual targeting signals in the
APP molecule (Anandatheerthavarada and others 2003).
These dual targeting signals of APP structurally share
homology with well-characterized chimeric signals
of CYP1A1, 2B1, and 2E1 (Addya and others 1997;
Anandatheerthavarada and others 1999; Robin and others
2002). Amino-terminal–35 amino acids constitute the
hydrophobic ER targeting signal, followed by the mito-
chondrial targeting signal encompassing between 35 and
67 amino acids (Fig. 4). APP occurs in three major iso-
forms as a result of alternative splicing of the gene (Price
and Sisodia 1998). The APP gene contains 19 exons.
Exon 7 encodes the Kunitz-type protease-inhibitor (KPI)
domain of 57 amino acids, and exon 8 encodes the 19-
amino-acid–long MRC OX-2 domain (Fig. 5A). APP770,
a non-neuronal form, contains KPI and OX-2 domains,
whereas APP751, another non-neuronal form, contains
only the KPI domain. However, the shorter neuron-
specific APP695 is devoid of KPI and OX-2 domains.
Interestingly,some major domains such as the N-terminal
dual targeting domain,the internal domain (180–290) rich
in acidic amino acids, and the C-terminal Aβ domain are
conserved in all three forms of APP (Fig. 5A).


________
Fig. 5. Domains in full-length amyloid precursor protein (APP). (A) Cartoon depicting the various domains of full-length APP770 isoform (complete gene product). ER TD =endoplasmic reticulum targeting domain; Mito TD =mitochondrial targeting domain; Aβ=β-amyloid peptide. (B) Mutations on APP around the Aβ domain have been identified in a number of families. These mutations are named after the affected families that are associated with early occurrence of Alzheimer’s disease.
________



APP is significantly modified by N- and O-glycosylation
during transport from the ER to the plasma membrane
via golgi (Annaert and others 1999). Most of the N-
terminuses of APP associated with the plasma membrane
orient toward the extracellular matrix. APP undergoes C-
terminal proteolytic cleavage, which is executed by α-,
β-, and γ-secretases (Nunan and Small 2000). First, β-
secretase cleaves APP at the C-terminus to generate a
99–amino-acid fragment (C99). Following this,the action
of γ-secretase on C99 produces two major Aβ forms,
namely Aβ40 and Aβ42. It is thought that Aβ42 is pro-
duced in large amounts in AD-affected brains (Tabira and
others 2002; Borchelt and others 1996). A number of
mutations in the APP gene have been discovered in and
around the Aβ region in the familial form of AD (Fig. 5B).
These mutations in the APP gene are associated with
increased production of Aβ42, thus causing early onset of
familial AD.


APP Expression and Mitochondrial Abnormalities in Cellular Models

In the past, studies have suggested a link between
APP overexpression and mitochondrial dysfunction.
Subsequently, mitochondrial structural and functional
changes were observed in cultured human muscle-fiber
cells and mouse embryonal carcinoma (P19) cells over
expressing human APP751 (Askanas and others 1996;
Grant and others 1999). In addition,association between
mitochondrial dysfunction and the intracellular accumula-
tion of C-terminal fragments of APP in cellular models
and in the brains of Down syndrome patients has been
suggested (Busciglio and others 2002). Furthermore,APP
is up-regulated under deprivation of tropic factors and by
aging (Stephenson and others 1992; Shepherd and others
2000; Bahmanyar and others 1987; Jeong and others
1997). The threshold expression between the wild-type
and familial mutant APP needed to cause cellular abnor-
mality is variable among cellular and mouse models of
AD (Janus and Westaway 2001; Hashimoto and others
2000). These studies collectively suggested that increased
expression of APP might be one of the causative factors
involved in mitochondrial dysfunction. On the other hand,
these studies did not address whether APP/Aβ can tar-
get mitochondria and directly affect mitochondrial vital
functions, and if so, what the AD relevance of APP/Aβ-
mediated mitochondrial abnormalities is.


Characteristics of Mitochondrial Translocation of APP in Cellular Cultures and Animal Models of AD

Using in vitro mitochondrial-import and in vivo neuronal-
expression studies,we showed that the endogenous as well
as the ectopically expressed Alzheimer’s full-length wild-
type and Swedish APP695 are localized to both the plasma
membrane and the mitochondria of human HCN-1A neu-
rons (Anandatheerthavarada and others 2003). These
human HCN neurons are non-transformed and offer
advantages to study various neuronal functions as they
show neurochemical and morphological characteristics of
primary cortical neuronal cells (Ronnet and others 1990).
Mutational analysis has identified that thepositively
charged residues at 40, 44, and 51 of the N-terminus of
APP are important components of the mitochondrial tar-
geting signal (Fig. 4). Subsequent studies also confirmed
the mitochondrial localization of full-length APP in PC12
cells and HEK293 cells stably transfected with Swedish
APP751 and APP695, respectively (Kiel and others 2004;
Park and others 2006).

Mitochondrially located APP differs from its plasma-
membrane counterpart with respect to glycosylation sta-
tus. Mitochondrial APP is a non-glycosylated protein, as
opposed to plasma-membrane APP, because glycosidase
treatment had no effect on the mobility of mitochondrial
APP. As evidenced by limited trypsin treatment, the
orientation of mitochondrial APP is such that the NH 2-
terminal resides inside the mitochondria,whereas most of
the COOH-terminalof the protein faces the cytosolic side
(Fig. 6Aand 6C). The orientation of mitochondrial APP
was further confirmed by multiple biochemical and cell
biological techniques (Park and others 2006). In vitro
chemical cross-linking studies and the immunoelectron
microscopy technique suggested that mitochondrially
localized APP was in contact with mitochondrial inner
and outer translocase proteins,suggesting the stable inter-
action between APP and mitochondrial translocases
(Anandatheerthavarada and others 2003; Fig. 6A and 6C).
This is in contrast to other mitochondrial proteins whose
interactions with import receptors under normal physio-
logical conditions are transient (Truscot and others 2001;
Rehling and others 2001). Further experiments involving
deletion mutants showed that the acidic domain spanning
220 to 290 amino acids of APP may be responsible for
the incomplete translocation, as deletion of this domain
(∆220–290) resulted in the complete mitochondrial
translocation of mutant APP (Fig. 6B). It is also
observed that with time,ectopically expressed wild-type
and Swedish APP accumulated on mitochondria of
human HCN-1A neurons and PC12 cells, respectively
(Anandatheerthavarada and others 2003; Keil and others
2004). Similarly, the mouse model of AD (Tg2576) that
overexpresses Swedish APP also showed the accumula-
tion of incompletely translocated full-length APP in the
mitochondrial compartment of the cortex and hippocam-
pus known to be affected in AD (Anandatheerthavarada
and others 2003).


________
Fig. 6. Models for formation of translocational intermediates mediated by the acidic domain of mitochondrial amyloid precursor protein (APP): Models Aand Cshow the N(mito)–C(cyto) orientation of APP resulting from the incomplete translocation caused by the acidic domain. (A) During the mitochondrial targeting of APP, the acidic domain may form a bulky structure because of increased intracellular metal-ion concentration or the inability of cytosolic chaperones to maintain the acidic domain in an unfolded manner. Depending on the bulkiness of the acidic domain, APP may not completely translocate through outer-membrane import TOM40 channels. (B) Complete mitochondrial translocation of deletion- mutant APP (∆220–290 APP) occurs when the internal acidic domain is removed, suggesting the role of the acidic domain in the translocational arrest of APP. (C) The negative charges around the APP acidic domain may experience electrostatic repulsion with highly negatively charged TOM22, an important component of outer-membrane protein-import machinery. This may lead to incomplete translocation of APP in the TOM40 channel. OM =outer membrane; TOM =translocase of the outer membrane; IMS =intermembrane space; IM =inner membrane; N =amino terminus; C = carboxy terminus.
________



Incomplete Mitochondrial Translocation of APP in Postmortem Human AD Brains

A recent study from our laboratory extensively exam-
ined the AD relevance of mitochondrial accumulation
of APP, using clinically diagnosed postmortem sporadic
AD (n=20) and non-AD (n=20) brains (Devi and
others 2006). Based on National Institutes of Health
(NIH)–Reagan criteria,the severity of AD was classified
into mild, moderate, and high categories. A significant
amount of full-length and C-terminal–truncated (lacking
Aβ domain) APP species accumulated in the mitochon-
dria of all three categories of AD brains as compared to
age-matched control brains,which showed very low lev-
els of mitochondrial APP. Nevertheless, APP levels in
human mitochondria of AD brains varied from 0.1 μg to
2.5 μg APP/mg mitochondrial protein, depending on the
severity of the disease. In support of neuronal culture
data, in human AD patients, mitochondrial-associated
APP is a non-glycosylated protein and is closely associ-
ated with outer-membrane channel-forming TOM40 in
an N(in mito)–C(out cyto) manner, suggesting the incomplete
translocation of APP.

Pathogenesis of AD is primarily associated with selec-
tive degeneration of specific brain regions and neuronal
types. To gain insight into AD relevance of regional and
cellular distribution of mitochondrial APP in AD brains,
we estimated mitochondrial APP levels in 11 different
brain regions of all three categories of AD. Interestingly,
mitochondrial APP levels in AD brains varied from
region to region. However, the frontal cortex, hippocam-
pus, and amygdala showed the highest accumulation of
APP in the mitochondria of all three categories of AD
brains. Triple-labeling immunohistochemistry of AD
brains revealed the accumulation of APP in the mito-
chondria of cholinergic neurons of all stages of AD
brains. Surprisingly, severe AD cases also showed the
accumulation of APP in the mitochondria of dopaminer-
gic, GABAergic, and glutamatergic neurons (Devi and
others 2006).


Interactions of Mitochondrially Accumulated APP with Import-Channel Proteins in the AD Brain

That incomplete translocation of APP occurs in the
mitochondria of the human AD brain was further sup-
ported by blue-native gel electrophoresis coupled with
Western blot analysis, which identified APP associated
with two ~480 and ~620 kDa complexes of high molec-
ular weight. Further analysis with antibodies specific to
the outer- and inner-membrane translocases revealed
that the ~480 kDa complex represents translocationally
arrested APP across the TOM40 complex, whereas the
~620 kDa complex represents APP arrested across both
the TOM40 and the TIM23 complexes. These data sug-
gest the possibility of formation of at least two different
steady-state import intermediates in the mitochondria of
the AD brain. Furthermore,higher amounts of APP were
found across mitochondrial import channels in severe
cases of AD. Taken together, culture and human data
suggest that translocationally arrested APP caused by
the acidic domain accumulates in physiologically impor-
tant channels in AD brains (Fig. 6).


Progressive Accumulation of APP across Import Channels and Mitochondrial Dysfunction

Studies have found that decreased mitochondrial functions
result from mitochondrial accumulation of full-length APP
in various cell models (Anandatheerthavarada and others
2003; Devi and others 2006; Keil and others 2004; Park
and others 2006; Fig. 7A). In HCN neuronal cells, mito-
chondrial accumulation of APP695 was associated with
decreased mitochondrial membrane potential,ATP levels,
and mitochondrial cytochrome coxidase activity that is
known to be decreased in AD (Anandatheerthavarada and
others 2003). However,accumulationof APP that lacks the
acidic domain showed no decreased cytochrome coxidase
or impaired energy metabolism,suggesting that the incom-
plete mitochondrial translocation of APP mediated by the
acidic domain might be the cause for the observed mito-
chondrial dysfunction (Anandatheerthavarada and others
2003). Similarly,decreased cytochrome c oxidase activity,
decreased ATP levels,and increased nitric oxide (NO) lev-
els were also associated with the mitochondrial accumula-
tion of full-length KPI containing APP751 in PC12 cells
(Keil and others 2004).


________
Fig. 7. Mitochondrial dysfunction mediated by amyloid precursor protein (APP) and β-amyloid (Aβ) during the pathogene- ig. 7. Mitochondrial dysfunction mediated by amyloid precursor protein (APP) and β-amyloid (Aβ) during the pathogenesis of Alzheimer’s disease. (A) Mitochondrial accumulation of full-length APP (translocational arrest) and Aβ are associated with impairment of a number of mitochondrial functions. Failure of mitochondrial functions ultimately may affect energy generation and increase oxidative stress. However, the status of proteases involved in preventing the accumulation of APP and Aβ in the mitochondrial compartment is not known under pathological conditions. Cytox =cytochrome c oxidase. (B) Under normal physiological conditions (normal cellular milieu and low levels of mitochondrial APP), the acidic domain of APP may form a folded structure, which is small enough to translocate through a large TOM40 channel (as compared to a TIM23 channel) successfully but big enough to get arrested in a small inner-membrane TIM23 channel. This may result in the access of a large portion of non–β-containing APP C-terminals to HtrA2 protease in the intermembrane space. Following the action of the protease, the C-terminal, 161–amino-acid, Aβ-containing APP fragment is released into the cytosol. OM =outer membrane; TOM =translocase of the outer membrane; TIM =translocase of the inner membrane; IMS =intermembrane space; IM =inner membrane; N =amino terminus; C =carboxy terminus.
________



It has been shown that the deficiency of nuclear-
coded cytochrome c oxidase IV and Vb subunits in
heart mitochondria during ischemia results in
decreased cytochromecoxidase activity and increased
oxidative stress (Prabu and others 2006). We tested
whether accumulation of APP in the mitochondrial
import channels of AD may block the GIP, which in
turn inhibits the mitochondrial entry of nuclear DNA–
coded cytochrome coxidase subunits IV and Vb (Devi
and others 2006; Fig. 7A). Ultimately, deficiency of
these important subunits may result in decreased
cytochrome c oxidase activity (Fig. 7A). Consistent
with this possibility, mitochondria isolated from AD
brains also showed a significant inhibition of import of
cytochrome coxidase subunits IV and Vb, which cor-
related with mitochondrial accumulation of APP.
Further analysis of all 20 AD brains showed impaired
cytochrome coxidase activity and increased H2O2 lev-
els. Interestingly, these mitochondrial abnormalities
directly correlated with the severity of AD (Devi and
others 2006; Fig. 7A).


Possible Factors Influencing the Translocational Arrest of APP during AD

The precise mechanism for translocational arrest mediated
by the acidic domain (220–290 amino acids) during AD is
not clear. However, several possibilities exist that may
occur during the pathogenesis of AD (Fig. 6). During the
pathogenesis of AD,cytoplasmic chaperones,which main-
tain the import-competent status of translocating proteins,
may not be able to fully unfold the polypeptide at the
acidic domain during mitochondrial import. This may
result in the formation of bulk folding of the acidic
domain. In addition to this, elevated levels of metal ions,
commonly observed in AD brains (Loeffler and others
1996; Basun and others 1991; Lovell and others 1998;
McDermott and others 1979), may increase the bulkiness
of the acidic domain by directly binding to it. Such bulki-
ness would inhibit APP translocation through the outer-
membrane TOM40 channel (Fig. 6A). Another possibility
may be that a negatively charged acidic domain may offer
electrostatic repulsion during the interaction with nega-
tively charged TOM22, an important component of the
TOM40 import-channel complex, resulting in the translo-
cational arrest of APP (Fig. 6C).


Accumulation of Aβ in Mitochondria in Cellular and Animal Models of AD

Besides accumulating full-length APP, the mitochondrial
compartment of neuronal cultures (Manczak and others
2006), AD transgenic mice (Caspersen and others 2005;
Manczak and others 2006; Crouch and others 2005) and
human AD brains (Lustbader and others 2004; Devi
and others 2006) also accumulate the C-terminal prote-
olytic product of APP, particularly the 4 kDa Aβ peptide
(Lustbader and others 2004; Caspersen and others 2005;
Manczak and others 2006; Crouch and others 2005; Fig.
7A). However, the association of Aβ immunoreactivity in
mitochondria was first observed in the AD brain by
immunoelectron microscopy (Yamaguchi and others
1992). Colocalization studies suggested that Aβ in the
mitochondrial compartment may be associated with the
soluble matrix (Caspersen and others 2005). Aβ that is
localized to mitochondria in the neurons of AD transgenic
mice and AD brains has been shown to inhibit the activity
of mitochondrial Aβ-binding alcohol dehydrogenase
(ABAD; Fig. 7A). In addition, X-ray crystallography
studies revealed that Aβ might act as a competitive
inhibitor by preventing the binding of nicotinamide ade-
nine dinucleotide to the active site of ABAD,thus leading
to mitochondrial oxidative damage (Lustbader and others
2004). A variety of biochemical techniques showed that
mitochondrial accumulation of Aβ in Tg2576 AD trans-
genic mice and mouse neuroblastoma cells expressing
human APP correlated with high levels of H2O2,
impaired cytochrome-oxidase activity, and increased
carbonylation of mitochondrial proteins (Manczak and
others 2006). In addition, the inhibitory effect of Aβ on
mitochondrial cytochrome c oxidase in the brains of
Tg2576 AD transgenic mice was accentuated by the pres-
ence of Cu2+ (Crouch and others 2005). Progressive accu-
mulation of Aβ in the mitochondria of TgmAPP (J-20
line), a transgenic mouse with targeted neuronal overex-
pression of mutant human APP, was associated with
impaired enzymatic activities of respiratory-chain com-
plexes III and IV as well as a reduced rate of mitochondr-
ial oxygen consumption (Caspersen and others 2005).
Taken together, these studies collectively suggest that Aβ
localizes to mitochondria and may impair some mito-
chondrial functions (Fig. 7A).


Aβ and Mitochondrial Dysfunction in Non-neuronal Cells

Recently, Aβ has been shown to induce mitochondrial
dysfunction in astrocytes and oligodendrocytes (Abramov
and others 2004; Xu and others 2001). Aβ-induced mito-
chondrial dysfunction in astrocytes was associated with
altered calcium signaling,free-radical generation,and the
activation of NADPH oxidase (Abramov and others
2004). Importantly, the survival of neurons following the
damage to the brain depends on glial cells, which secrete
several survival factors. Hence, the mitochondrial dys-
function in glial cells may have far-reaching implications
for neurotoxicity (Fig. 8). In support of this coculture sys-
tem consisting of astrocytes and neurons, Aβ-induced
mitochondrial dysfunction in astrocytes resulted in the
death of neurons (Abramov and others 2004). In addition,
it has been shown that Aβ can also deplete glutathione
levels in astrocytes. Based on this, it is hypothesized that
depletion of glutathione, an antioxidant, may result in
neuronal death because astrocytes supply amino-acid pre-
cursors for neuronal glutathione synthesis. Alternatively,
neuronal death may also be mediated through cytokines
released from dysfunctional astrocytes (Sue and Griffin
2006). These studies suggest that Aβ-induced mitochon-
drial dysfunction in astrocytes appears to affect their abil-
ity to maintain the integrity of neurons leading to
neurotoxicity (Fig. 8).


________
Fig. 8. Impact of Aβ-induced mitochondrial dysfunction in glial cells on neuronal survival. Glial dysfunction is one the g. 8. Impact of Aβ-induced mitochondrial dysfunction in glial cells on neuronal survival. Glial dysfunction is one the lesions observed in Alzheimer’s disease. Glial cells secrete many factors needed for neuronal survival. Accumulation of Aβ in the mitochondria of astrocytes may induce mitochondrial dysfunction, which in turn may affect the ability of astrocytes to secrete survival factors. Thus, deprivation of neurons with survival factors may result in neuronal toxicity and possibly death.
________



Genesis of Mitochondrial Aβ

The origin of mitochondrial Aβ is not known. Furthermore,
whether the genesis of mitochondrial Aβ is an early or late
event in the pathogenesis of AD also is not clear. The pres-
ence of mitochondrial APP and the discovery of a γ-
secretase such as protease in mitochondria may raise
the possibility that Aβ is being produced in the mitochon-
drial compartment (Hansson and others 2004). However,
the N(mito) and C(cyto) orientation of mitochondrial APP
(Anandatheerthavarada and others 2003; Devi and others
2006; Park and others 2006) along with the intramito-
chondrial nature of γ-secretase may not support the possi-
bility of Aβ generation inside the mitochondria.
Additionally, the presence of β-secretase—like activity in
the mitochondrial fraction,which is important for the gen-
eration of substrate peptide C99 for γ-secretase—is not
known. On the other hand, simultaneous presence of C-
terminal–truncated (lacking Aβ) APP and the Aβ in the
mitochondria of the AD brain may suggest a couple of
possibilities for the genesis of extramitochondrial Aβ.
One possibility may be that with the progression of AD,
there is a loss of the cytosolically exposed,Aβ-containing
C-terminus of mitochondria-associated APP by some
unknown mechanisms, thus generating the Aβ species
outside the mitochondria. Alternatively,during mitochon-
drial import, the Aβ-containing C-terminus of APP may
be cleaved in the cytosol to generate extramitochondrial
Aβ species that may be transported into mitochondria by
an unknown mechanism.


Turnover/Degradation of Mitochondrial APP and Aβ

Turnover and degradation of APP and Aβ in the mito-
chondrial compartment appear to be important, because
the mitochondrial accumulation of full-length APP and
Aβ plays a key role in mitochondrial dysfunction.
Recently, at least three mitochondrial resident proteases
involved in the turnover of APP and Aβ have been
reported (Park and others 2006; Leissring and others
2004; Falkevall and others 2006).

Aβ-Degrading Mitochondrial Proteases

A novel mitochondrial protease called presequence pepti-
dase (PreP),which belongs to the pitrilysin oligopeptidase
family M16C and contains an inverted zinc-binding motif,
is found in the brain’s mitochondrial matrix (Falkevall and
others 2006; Fig. 7A). This protease is capable of degrad-
ing presequences and short peptides. Degradation of syn-
thetic Aβ1-42 and 1-40 by recombinant human PreP
under in vitro conditions has been demonstrated (Falkevall
and others 2006). Another study has identified a mito-
chondrial isoform of insulin-degrading enzyme (IDE).
IDE is a widely expressed extracellular zinc metallopepti-
dase and is capable of degrading both cerebral amyloid
β-peptide and plasma insulin (Leissring and others 2004).
Interestingly, the mitochondrial isoform is generated by
alternate translation, which results in the addition of a 41–
amino-acid N-terminal mitochondrial targeting sequence
(Leissring and others 2004). However, the physiological
and pathological roles of the newly identified mitochondrial
IDE and PreP with respect to the turnover of mitochondrial
Aβ have not yet been demonstrated.

APP-Degrading Mitochondrial Protease

Using mutational studies under in vitro and in vivo con-
ditions, it has recently been shown that serine protease
HtrA2, which localizes to mitochondrial intermembrane
space,cleaves mitochondrial APP at amino acid 535 and
releases C-terminal Aβ containing a fragment 161 amino
acids long into the cytosol (Park and others 2006; Fig.
7B). Furthermore, this process occurs under normal
physiological conditions, thus preventing the mitochon-
drial accumulation of APP. This supports the observation
that decreased turnover of mitochondrial APP may result
in the time-dependent accumulation of APP across mito-
chondrial membranes, which may lead to mitochondrial
dysfunction (Anandatheerthavarada and others 2003).
It is not clear how the protease gains access to the C-
terminal 535th residue of APP in the intramembranous
space. However, based on our studies in cell culture and
human postmortem AD brains, it is possible that the
incomplete translocation of APP may form several
translocation intermediates, depending on the cellular
conditions spanning outer- and/or inner-membrane
import channels (Anandatheerthavarada and others
2003; Devi and others 2006). Under normal physiologi-
cal conditions, there may be a possibility that APP mol-
ecules may cross the outer membrane TOM40 channel
and get arrested in the inner membrane TIM23 pore,
which is believed to be smaller than the TOM40 pore. This
may provide access to HtrA2 to cleave the C-terminal
535th residue of APP in the intramembranous space,and
thus,the cleaved Aβ-containing C-terminal fragment may
be released to the cytosol (Fig. 7B).


Implications of APP/Aβ-Mediated Mitochondrial Dysfunction in Non-cognitive Neuropsychiatric Symptoms of AD

The brain is a complex organ with much cellular,regional,
and functional heterogeneity. Several neurotransmitter sys-
tems have been shown to be perturbed in the AD brain
(Reinikainen and others 1990; Ellison and others 1986;
Nazarali and Reynolds 1992). In addition to an impairment
of cholinergic neuronal-mediated cognitive functions
in AD patients, non-cognitive neuropsychiatry symp-
toms such as depression and psychosis also exist and
are believed to be associated with perturbation of non-
cholinergic neurotransmission (Mohs 1996; Zubenko
2000; Cummings 2000). The literature reviewed above
suggests that the accumulation of full-length APP and Aβ
in the mitochondrial compartment may play an important
role in causing mitochondrial dysfunction (Fig. 7A). Taken
together, APP/Aβ-mediated mitochondrial dysfunction in
various neuronal systems in different brain regions may
have far-reaching influence on neuronal survival and on
AD-specific cognitive and non-cognitive neuropsychiatric
behavior.

At this juncture, it is thought that mitochondria are
direct targets for full-length APP and Aβ (Fig. 7A
and 8). Future studies are needed to understand how the
coordination between translocationally arrested mito-
chondrial full-length APP and mitochondrial Aβ causes
the impairment of mitochondrial vital functions in the
pathogenesis of AD.


References

Abramov AY, Canevari L, Duchen MR. 2004. βAmyloid peptides
induce mitochondrial dysfunction and oxidative stress in astrocytes
and death of neurons through activation of NADPH oxidase. J
Neurosci 24:565-75.

Addya S,Anandatheerthavarada HK,Biswas G,Bhagwat SV,Mullick J,
Avadhani NG. 1997. Targeting of NH
2-terminal–processed microso-
mal protein to mitochondria: a novel pathway for the biogenesis of
hepatic mitochondrial P450MT2. J Cell Biol 139:589-99.

Anandatheerthavarada HK, Biswas G, Mullick J, Sepuri NB, Otvos L,
Pain D,and others. 1999. Dual targeting of cytochrome P4502B1 to
endoplasmic reticulum and mitochondria involves novel signal acti-
vation by cyclic AMP-dependent phosphorylation at ser128. EMBO
J 18:5494-504.

Anandatheerthavarada HK, Biswas G, Robin MA, Avadhani NG.
2003. Mitochondrial targeting and a novel transmembrane arrest of
Alzheimer’s amyloid precursor protein impairs mitochondrial
function in neuronal cells. J Cell Biol 161:41-54.

Annaert WG, Levesque L, Craessaerts K, Dierinck I, Snellings G,
Westaway D, and others. 1999. Presenilin 1 controls gamma-
secretase processing of amyloid precursor protein in pre-golgi
compartments of hippocampal neurons. J Cell Biol 147:277-94.

Askanas V, McFerrin J, Baque S, Alvarez RB, Sarkozi E, Engel WK.
1996. Transfer of beta-amyloid precursor protein gene using aden-
ovirus vector causes mitochondrial abnormalities in cultured nor-
mal human muscle. Proc Natl Acad Sci USA 93:1314-9.

Bahmanyar S, Higgins GA, Goldgaber D, Lewis DA, Morrison JH,
Wilson MC,and others. 1987. Localization of amyloid beta protein
messenger RNA in brains from patients with Alzheimer’s disease.
Science 237:77-80.

Baloyannis SJ. 2006. Mitochondrial alterations in Alzheimer’s disease.
J Alzheimer’s Dis 9:119-26.
Basun H, Forssell LG, Wetterberg L, Winblad B. 1991. Metals and
trace elements in plasma and cerebrospinal fluid in normal aging
and Alzheimer’s disease. J Neural Transm Park Dis Dement Sect
3:231-58.

Beal MF. 1998. Mitochondrial dysfunction in neurodegenerative dis-
eases. Biochim Biophys Acta 1366:211-23.

Beal MF. 2005. Mitochondria take center stage in aging and neurode-
generation. Ann Neurol 58:495-505.

Borchelt DR, Thinakaran G, Eckman CB, Lee MK, Davenport F,
Ratovitsky T, and others. 1996. Familial Alzheimer’s disease-
linked presenilin 1 variants elevate Abeta1-42/1-40 ratio in vitro
and in vivo. Neuron 17:1005-13.

Busciglio J, Pelsman A,Wong C, Pigino G,Yuan M, Mori H, and oth-
ers. 2002. Altered metabolism of the amyloid beta precursor pro-
tein is associated with mitochondrial dysfunction in Down’s
syndrome. Neuron 33:677-88.

Casley CS, Canevari L, Land JM, Clark JB, Sharpe MA. 2002. Beta-
amyloid inhibits integrated mitochondrial respiration and key
enzyme activities. J Neurochem 80:89-100.

Caspersen C,Wang N,Yao J,Sosunov A,Chen X,Lustbader JW,and oth-
ers. 2005. Mitochondrial Aβ: a potential focal point for neuronal
metabolic dysfunction in Alzheimer’s disease. FASEB J 19:2040-1.

Chang S,Ma T,Miranda D,Balestra ME,Mahley RW,Huang Y. 2005.
Lipid- and receptor-binding domains of apolipoprotein E4 frag-
ments act in concert to cause mitochondrial dysfunction and neu-
rotoxicity. Proc Natl Acad Sci USA 102:18694-9.

Chen H, Chan DC. 2005. Emerging functions of mammalian mito-
chondrial fusion and fission. Human Mol Gen 15(2):R283-9.

Crouch PJ,Blake R,Duce JA,Ciccotosto GD,Li QX,Barnham KJ,and
others. 2005. Copper-dependent inhibition of human cytochrome
coxidase by a dimeric conformer of amyloid-beta1–42. J Neurosci
25:672-9.

Cummings JL. 2000. Cognitive and behavioral heterogeneity in
Alzheimer’s disease: seeking the neurobiological basis. Neurobiol
Aging 21:845-61.

David DC, Hauptmann S, Scherping I, Schuessel K, Keil U, Rizzu P,
and others. 2005. Proteomic and functional analysis reveal a mito-
chondrial dysfunction in P301L Tau transgenic mice. J Biol Chem
280:23802-14.

Devi L, Prabhu BM, Galati DF,Avadhani NG,Anandatheerthavarada
HK. 2006. Accumulation of amyloid precursor protein in the
mitochondrial import channels of human Alzheimer’s disease
brain is associated with mitochondrial dysfunction. J Neurosci
30:9057-68.

Ellison DW,Beal MF,Mazurek MF,Bird ED,Martin JB. 1986. A post-
mortem study of amino acid neurotransmitters in Alzheimer’s dis-
ease. Ann Neurol 20:616-21.

Endo T,Yamamoto H,Esaki M. 2003. Functional cooperation and sep-
aration of translocators in protein import into mitochondria, the
double-membrane bounded organelles. J Cell Sci 116:3259-67.

Falkevall A,Alikhani N, Bhushan S, Pavlov PF, Busch K, Johnson KA,
and others. 2006. Degradation of the amyloid beta protein by the
novel mitochondrial peptidasome,PreP*. Biol Chem 281:29096-104.

Gabaldon T, Huynen MA. 2004. Shaping the mitochondrial proteome.
Biochem Biophys Acta 1659:212-20.

Gouras GK,Almeida CG,Takahashi RH. 2005. Intraneuronal Abeta accu-
mulation and origin of plaques in Alzheimer’s disease. Neurobiol
Aging 26:1235-44.

Grant SM, Shankar SL, Chalmers-Redman RM, Tatton WG, Szyf M,
Cuello AC. 1999. Mitochondrial abnormalities in neuroectodermal
cells stably expressing human amyloid precursor protein
(hAPP751). Neuroreport 18:41-6.

Hansson CA, Frykman S, Farmery MR, Tjernberg LO, Nilsberth C,
Pursglove SE, and others. 2004. Nicastrin, presenilin,APH-1, and
PEN-2 form active gamma-secretase complexes in mitochondria. J
Biol Chem 279:51654-60.

Hashimoto Y, Niikura T, Ito Y, Nishimoto I. 2000. Multiple mechanisms
underlie neurotoxicity by different types of Alzheimer’s disease
mutations of amyloid precursor protein. J Biol Chem 275:34541-51.

Janus C,Westaway D. 2001. Transgenic mouse models of Alzheimer’s
disease. Physiol Behav 73:873-86.

Jeong SJ, Kim K, Suh YH. 1997. Age-related changes in the expres-
sion of Alzheimer’s beta APP in the brain of senescence acceler-
ated mouse (SAM)-P/10. Neuroreport 8:1733-7.

Keil U, Bonert A, Marques CA, Scherping I, Weyermann J,
Strosznajder JB, and others. 2004. Amyloid beta-induced changes
in nitric oxide production and mitochondrial activity lead to apop-
tosis. J Biol Chem 279:50310-20.

Leissring MA, Farris W, Wu X, Christodoulou DC, Haigis MC,
Guarente L, and others. 2004. Alternative translation initiation
generates a novel isoform of insulin-degrading enzyme targeted to
mitochondria. Biochem J 383:439-46.

Lin MT, Beal MF. 2006. Mitochondrial dysfunction and oxidative
stress in neurodegenerative diseases. Nature 443:787-95.

Loeffler DA, LeWitt PA, Juneau PL, Sima AA, Nguyen HU, DeMaggio
AJ,and others. 1996. Increased regional brain concentrations of ceru-
loplasmin in neurodegenerative disorders. Brain Res 738:265-74.

Lovell MA, Robertson JD, Teesdale WJ, Campbell JL, Markesbery
WR. 1998. Copper, iron and zinc in Alzheimer’s disease senile
plaques. J Neurol Sci 158:47-52.

Lustbader JW, Cirilli M, Lin C, Xu HW,Takuma K,Wang N, and oth-
ers. 2004. ABAD directly links Abeta to mitochondrial toxicity in
Alzheimer’s disease. Science 304:448-52.

Manczak M, Anekonda TS, Henson E, Park BS, Quinn J, Reddy PH.
2006. Mitochondria are a direct site of A beta accumulation in
Alzheimer’s disease neurons: implications for free radical genera-
tion and oxidative damage in disease progression. Hum Mol Genet
15:1437-49.

Manfredi G, Beal MF. 2000. The role of mitochondria in the patho-
genesis of neurodegenerative diseases. Brain Pathol 10:462-72.

Mattson MP. 1997. Central role of oxyradicals in the mechanism of
Amyloid β-peptide toxicity. Alzheimer’s Disease Review 2:1-14.

Maurer I, Zierz S, Moller HJ. 2000. A selective defect of cytochrome
c oxidase is present in brain of Alzheimer disease patients.
Neurobiol Aging 21:455-62.

McDermott JR, Smith AI, Iqbal K, Wisniewski HM. 1979. Brain alu-
minum in aging and Alzheimer disease. Neurology 29:809-14.

Mohs RC. 1996. The Alzheimer’s disease assessment scale. Int
Psychogeriatr 8:195-203.

Nazarali AJ, Reynolds GP. 1992. Monoamine neurotransmitters and
their metabolites in brain regions in Alzheimer’s disease: a post-
mortem study. Cell Mol Neurobiol 12:581-7.

Nicholls DJ, Budd SL. 2000. Mitochondria and neuronal survival.
Physiol Rev 80(1):315-60.

Nunan J,Small DH. 2000. Regulation of APP cleavage by alpha-,beta-
and gamma-secretases. FEBS Letters 24184:1-5

Offen D, Elkon H, Melamed E. 2000. Apoptosis as a general cell death
pathway in neurodegenerative diseases. J Neural Transm 58:153-66.

Park HJ,Kim SS,Seong YM,Kim KH,Goo HG,Yoon EJ,and others.
2006. Amyloid precursor protein is a direct cleavage target of
HtrA2 serine protease. J Biol Chem 281:34277-87.

Parker WD Jr. 1991. Cytochrome oxidase deficiency in Alzheimer’s
disease. Ann NY Acad Sci 640:59-64.

Prabu SK,Anandatheerthavarada HK, Raza H, Srinivasan S, Spear JF,
Avadhani NG. 2006. Protein kinase A-mediated phosphorylation
modulates cytochrome coxidase function and augments hypoxia
and myocardial ischemia-related injury. J Biol Chem 281:2061-70.

Price DL, Sisodia SS. 1998. Mutant genes in familial Alzheimer’s dis-
ease and transgenic models. Annu Rev Neurosci 21:479-505.

Reddy PH. 2006. Amyloid precursor protein-mediated free radicals
and oxidative damage: implications for the development and
progression of Alzheimer’s disease. J Neurochem 96:1-13.

Reddy PH, Beal MF. 2005. Are mitochondria critical in the patho-
genesis of Alzheimer’s disease? Brain Res Brain Res Rev 49:
618-32.

Rehling P, Wiedemann N, Pfanner N, Truscott KN. 2001. The mito-
chondrial import machinery for preproteins. Crit Rev Biochem
Mol Biol 36:291-336.

Reinikainen KJ, Soininen H, Riekkinen PJ. 1990. Neurotransmitter
changes in Alzheimer’s disease: implications to diagnostics and
therapy. J Neurosci Res 27:576-86.

Robin MA, Anandatheerthavarada HK, Biswas G, Naresh Babu VS,
Gordon DM,Pain D,and others. 2002. Bimodal targeting of microso-
mal CYP2E1to mitochondria through activation of an N-terminal
chimeric signal by cAMP mediated phosphorylation. J Biol Chem
277:40583-93.

Roesch K, Curran SP, Tranebjaerg L, Koehler CM. 2002. Human
deafness dystonia syndrome is caused by a defect in assembly
of the DDP1/TIMM8a-TIMM13 complex. Hum Mol Genet 11:
477-86.

Ronnett GV,Hester LD,Nye JF,Conners K,Snyder SH. 1990. Human
cortical neuronal cell line: establishment from a patient with uni-
lateral megalencephaly. Science 248:603-5.

Selkoe DJ. 1999. Translating cell biology into therapeutic advances in
Alzheimer’s disease. Nature 399:23-31.
Shepherd CE, Bowes S, Parkinson D, Cambray-Deakin M, Pearson
RC. 2000. Expression of amyloid precursor protein in human
astrocytes in vitro: isoform-specific increases following heat
shock. Neuroscience 99:317-25.

Sims NR. 1996. Energy metabolism, oxidative stress and neuronal
degeneration in Alzheimer’s disease. Neurodegeneration 5:435-40.

Stephenson DT, Rash K, Clemens JA. 1992. Amyloid precursor pro-
tein accumulates in regions of neurodegeneration following focal
cerebral ischemia in the rat. Brain Res 593:128-35.

Sue W, Griffin T. 2006. Inflammation and neurodegenerative diseases.
Am J Clin Nutr 83(suppl):470S–4S.

Tabira T,Chui DH,Kuroda S. 2002. Significance of intracellular Abeta42
accumulation in Alzheimer’s disease. Front Biosci 7:44-9.

Taylor SW, Fahy E, Zhang B, Glenn GM, Warnock DE, Wiley S, and
others. 2003. Characterization of the human heart mitochondrial
proteome. Nat Biotechnol 21:281-6.

Thinakaran GJ. 1999. The role of presenilins in Alzheimer’s disease.
Clin Invest 104:1321-7.

Truscott KN, Pfanner N, Voos W. 2001. Transport of proteins into
mitochondria. Rev Physiol Biochem Pharmacol 143:81-136.

Xu J, Chen S, Ahmed H, Chen H, Ku G, Goldberg MP, and others.
2001. Amyloid-β peptides are cytotoxic to oligodendrocytes. J
Neurosci 21:RC118:1-5.

Yamaguchi H,Yamazaki T, Ishiguro K, Shoji M, Nakazato Y, Hirai S.
1992. Ultrastructural localization of Alzheimer amyloid beta/A4
protein precursor in the cytoplasm of neurons and senile plaque-
associated astrocytes. Acta Neuropathol 85:15-22.

Zubenko GS. 2000. Neurobiology of major depression in Alzheimer’s
disease. Int Psychogeriatr 12:217-30.
________

The research work in the laboratory of the first author was supported by
National Institutes of Health/National Institute on Aging (NIH/NIA)
grant AG 021920. The authors thank National Disease Resources
Interchange (NDRI),Philadelphia,PA,for providing postmortem human
brains for the research work.

Address correspondence to: H. K. Anandatheerthavarada,Department of
Animal Biology,School of Veterinary Medicine,3800 Spruce Street,Room
#189E, University of Pennsylvania, Philadelphia, PA 19104 (e-mail:
ann1234@vet.upenn.edu).