|Home | About | Journals | Submit | Contact Us | Français|
Mitochondria are the major source of energy for the normal functioning of brain cells. Increasing evidence suggests that the amyloid precursor protein (APP) and amyloid beta (Aβ) accumulate in mitochondrial membranes, cause mitochondrial structural and functional damage, and prevent neurons from functioning normally. Oligomeric Ab is reported to induce intracellular Ca2+ levels and to promote the excess accumulation of intracellular Ca2+ into mitochondria, to induce the mitochondrial permeability transition pore to open, and to damage mitochondrial structure. Based on recent gene expression studies of APP transgenic mice and AD postmortem brains, and APP/Aβ and mitochondrial structural studies, we propose that the overexpression of APP and the increased production of Aβ may cause structural changes of mitochondria, including a increase in the production of defective mitochondria, a decrease in mitochondrial trafficking, and the alteration of mitochondrial dynamics in neurons affected by AD. This article discusses some critical issues of APP/Aβ associated with mitochondria, mitochondrial structural and functional damage, and abnormal intracellular calcium regulation in neurons from AD patients. This article also discusses the link between Aβ and impaired mitochondrial dynamics in AD.
Alzheimer’s disease (AD) is a late-onset, progressive, age-dependent neurodegenerative disorder, characterized clinically by the impairment of cognitive functions and changes in behavior and personality (Selkoe, 2001; Mattson, 2004; LaFerla et al., 2007; Reddy and Beal, 2008). AD is associated with intracellular neurofibrillary tangles and extracellular amyloid beta (Ab) plaques in the regions of the brain that are responsible for learning and memory. Recent molecular, cellular, gene expression and immunochemical studies of AD postmortem brains and brain specimens from AD transgenic mice revealed a loss of neuronal subpopulations, decreased presynaptic and postsynaptic immunoreactivity of terminals, a decrease in cholinergic fibers, the proliferation of reactive astrocytes and microglia, and abnormal mitochondrial structural and functional alterations (Mattson, 2004; LaFerla et al., 2007; Reddy 2006, 2007; Reddy and McWeeney, 2006; Reddy and Beal, 2005, Swerdlow, 2007). Among these important changes, mitochondrial oxidative damage and synaptic pathology have recently been reported as early events in AD progression (reviewed in Reddy and Beal, 2008; Pratico, 2008). However, it is still unclear the causal factors of mitochondrial oxidative damage and synaptic pathology in AD pathogenesis. Further, the precise link between Ab and the structure/function of mitochondria in the development and progression of AD - is not completely understood. This article outlines the role of the amyloid precursor protein (APP) and Ab in mitochondrial structural and functional alterations in neurons affected by AD. This article also discusses the latest developments in mitochondrial structural and functional abnormalities in relation to Ab in AD pathogenesis.
Recently, Aβ monomers and oligomers were found in mitochondrial membranes, in the neurons of postmortem brain specimens from AD patients and in neurons of brain specimens from AD mice and neuronal cells expressing mutant APP (Manczak et al., 2006; Crouch et al., 2005; Caspersen et al., 2006; Devi et al., 2006). However, the mechanism by which is Ab transported to mitochondrial membranes is not fully understood. Very recently, using immuno-electron microscopy, subcellular fractionation, and immunoblotting techniques, Hanson Peterson et al investigated the mitochondrial uptake of Ab in AD neurons. They found that Ab is transported into mitochondria via the translocase of the outer membrane machinery (Hanson Peterson et al., 2008). Subfractionation studies revealed that Ab is associated with the inner membrane of mitochondria, and immuno-electron microscopy studies showed Ab localized to the cristae of the inner membrane of mitochondria. Hanson Peters et al. also found a distribution pattern of Ab in the inner mitochondrial membrane in brain biopsies of the human cortex obtained from living subjects with normal-pressure hydrocephalus. Findings from the Hanson Peterson et al. study (2008), together those from earlier studies (Manczak et al., 2006; Crouch et al., 2005; Caspersen et al., 2006; Devi et al., 2006), suggest that Ab species transport to mitochondrial membranes (Fig. 1), cause mitochondrial dysfunction and oxidative damage, and ultimately damage AD neurons both structurally and functionally.
APP is reported to be associated with mitochondria in neurons affected by AD. Using in vitro (human cortical neurons or HCN-1A neurons) and in vivo (Tg2576 mice) studies, Anandatheerthavarada and colleagues researched the association of endogenous and ectopically expressed full-length, wild-type, and mutant APP. They found both wild-type and mutant APP (with Swedish mutation) localized to the plasma membrane and the mitochondria of human HCN-1A neurons (Anandatheerthavarada et al., 2003). Other groups also reported full-length wild-type and/or mutant APP localized to the mitochondria of PC12 cells and HEK293 cells that were stably transfected with Swedish APP751 and APP695 (Kiel et al., 2004; Park et al., 2006). Several groups demonstrated the localization of mitochondrial APP in the NH2 -terminal inside the mitochondria and COOH-terminal of the protein faces the cytosolic side (Anandatheerthavarada et al., 2003; Park et al 2006).
Using postmortem brain specimens from patients with mild, moderate, and severe AD, and from age-matched healthy control subjects, Devi and colleagues studied the association of APP with mitochondria. They found full-length APP and C-terminal–truncated (lacking Aβ domain) APP species accumulated in the mitochondria from all three categories of AD brains, as compared to the control brains. The accumulation of APP in the mitochondrial membranes increased with severity of AD, suggesting that APP associated with mitochondria may be critically involved in the disease process of AD (Fig. 1).
Mitochondrial dysfunction is a prominent feature of AD, but the underlying mechanism is still not completely understood. Mitochondrial dysfunction has been described in AD postmortem brains (Gibson et al., 1998; Maurer et al., 2000; Smith et al., 1996; Devi et al., 2006), platelets from AD patients (Parker et al., 1990), AD transgenic mice (Reddy et al., 2004; Lustbader et al., 2004; Li et al., 2004; Caspersen et al., 2005; Manczak et al., 2006; Eckert et al., 2008) and cell-lines that express mutant APP and/or cells treated with Ab (Kaminsky et al., 2008; Diana et al., 2008; Schmidt et al., 2008; Matsumoto et al., 2006 ).
Several lines of evidence suggest that mitochondrial abnormalities plays a large role in AD pathogenesis:
Structurally, mitochondria are compartmentalized with 2 lipid membranes: an inner membrane and an outer membrane. The inner membrane covers the mitochondrial matrix, which contains tricarboxylic acid and beta oxidation. The inner membrane houses the mitochondrial electron transport chain (ETC) and provides a highly efficient barrier to the flow of ions. The outer mitochondrial membrane is basically porous and allows the passage of low molecular-weight substances. Mitochondrial ATP is generated via oxidative phosphorylation. Through this succession of oxido-reduction reactions, electrons pass along the ETC complexes and generate an electrochemical gradient by fueling the extrusion of protons from the matrix, across the inner mitochondrial membrane complexes. ATP is then generated by the dissipation of this proton gradient through complex V. During the generation of ATP, ROS is generated as a physiological by-product of the ETC chain. During the transfer of electrons to molecular oxygen, 0.4% to 5% of ETC lose their way and participate in the formation of superoxide radicals (O2•-) (Reddy, 2008). The increase in O2•- generated from this loss of electrons ultimately activates the mitochondrial permeability transition pore, causes damage to the structure of mitochondria, and ultimately destroys the cell by apoptosis.
Several lines of evidence suggest that increased mitochondrially generated free radicals are responsible for changes in mitochondrial morphology, including mitochondrial fission and fusion imbalance (Fig. 2): 1. Barsoum et al. (2006) found that mitochondria undergo mitochondrial fission when treated with Ab in AD. 2. Bernard et al (2006) also found that when mammalian cells were treated with rotenone (pesticide), mitochondrial ROS production increased and cellular ATP production decreased, with these changes ultimately leading to mitochondrial fragmentation. 3. Yoon et al (2006) studied the link between mitochondrial fission and the high glucose-induced overproduction of ROS. They found that mitochondria are fragmented, with a concomitant increase in ROS after the exposure of cells to high glucose concentrations, suggesting that the dynamic change of mitochondrial morphology in high-glucose conditions contributes to ROS overproduction (Yoon et al., 2006). It is possible that in late-onset neurodegenerative diseases, such as AD, an increase in age-related ROS production may cause mitochondrial fragmentation, which may lead to mitochondrial fission and fusion imbalance, mitochondrial dysfunction, and neuronal damage.
Based on the findings of our laboratory and others, we hypothesize that in AD, Ab is transported to mitochondria, which induces free radicals, causes mitochondrial structural and functional alterations, and ultimately damages neurons. To support this hypothesis, in a recent study using confocal and electron microscopy and M17 cells transfected with wild-type or mutant APP, Wang et al investigated the effect of APP and Ab on mitochondrial dynamics in neurons. Confocal and electron microscopic analysis demonstrated that about 40% of M17 cells overexpressing wild-type APP (APPwt M17 cells) and more than 80% of M17 cells overexpressing mutant APP (APPswe M17 cells) displayed alterations in mitochondrial morphology, particularly fragmented mitochondria. They also found increased levels of Fis1 is critical for mitochondrial fission in APPwt and APPswe M17 cells. The overexpression of APP and/or Ab -derived diffusible ligand treatment also led to mitochondrial fragmentation and reduced mitochondrial coverage in neuronal processes in differentiated primary hippocampal neurons.
In another study, Du et al (2008) investigated the connection between Ab and Cyclophilin D (a mitochondrial matrix protein) using AD postmortem brains, primary neuronal cultures from CypD knockout mice, APP transgenic mice, and double mutant mice (APP transgenic and Cyclophilin D deficient mice). Using cell biology, and electron and confocal microscopy techniques, they found that Aβ interacts with the mitochondrial matrix protein Cyclophilin D. To further investigate the interaction between Aβ and CypD, Du et al (2008) crossed Cyclophilin D knockout mice with APP transgenic mice and investigated Aβ pathology, Cyclophilin D expression, and the cognitive behavior of Cyclophilin D knockout mice, APP transgenic mice, and double-mutant mice (APP transgenic and Cyclophilin D knockout mice). They found that a deficiency in Cyclophilin D attenuated Aβ-induced mitochondrial oxidative stress, improved synaptic function, and ameliorated cognitive deficits in the double-mutant mice (Du et al., 2008). These findings suggest that a decreased interaction of Aβ with Cyclophilin D prevents the mitochondrial permeability transition pore from opening and preserves mitochondrial structure and function .
Based on findings about Aβ and mitochondria, we propose a model of mitochondrial dynamics in AD neurons (Fig. 3) in which increased APP and Aβ in association with mitochondria – mitochondrial trafficking and mitochondrial dynamics – are decreased in neurons from AD patients compared to non-demented healthy subjects. In other words, in this model of mitochondrial dynamics, increased mitochondrial APP and Aβ cause increase mitochondrial fragmentation and decrease the number of functionally active mitochondria, which ultimately imbalances mitochondrial dynamics in neurons from AD patients.
Abnormal homeostasis of intracellular Ca2+ levels has been observed in aging and AD (Brown et al 2004, 2006; Naga et al 2008; Bezprozvanny and Mattson 2008; Giasomello et al., 2007; Norenberg and Rao, 2007; Mattson, 2007; Stavrovskaya and Kristal, 2005; Brzyska and Elbaum 2003; Huang et al., 2003; Chan et al., 2002; LaFerla, 2002, Pascale and Etcheberrigaray, 1999). Several lines of evidence support the involvement of calcium dyshomeostasis in AD. Mutant presenilins are known to increase the intracellular Ca2+ levels in AD neurons, and promote the entry of Ca2+ levels into mitochondria (Leissring et al., 2000; Chan et al., 2000; Begley et al., 1999; Guo et al., 1998; Furukawa et al., 1998), and Aβ has been reported to promote Ca2+ entry into neurons and neuronal mitochondria (Green and LaFerla, 2008; Sanz-Blasco et al., 2008; Mattson, 1994). It was unclear until recently which form of Ab – monomeric, oligomeric, or protofibrils – promotes the entry of Ca2+ into neurons and neuronal mitochondria. Oligomers of Aβ have been reported to be more toxic to neurons than other forms of Ab . However, the link between putative changes in intracellular Ca2+ and cell damage is still not completely understood. As discussed above, increased intramitochondrial Ca2+ may help damage the inner mitochondrial membrane and may lead to structural and functional abnormalities of mitochondria (Fig. 4).
Recently, using photon-counting imaging of neurons expressing targeted aequorin, Nunez et al (2007) addressed the effects of Aβ assembly state on Ca2+ influx and mitochondrial Ca2+ uptake (Nunez et al. 2007). Sanz-Blasco et al (2008) investigated the connection between oligomers, and intracellular and intramitochondrial calcium levels using cerebellar granule cells and Ab oligomers (Sanz-Blasco et al 2008). They found that only Ab1–42 oligomers induced a massive entry of Ca2+ into neurons and promoted mitochondrial Ca2+ overload. Ab oligomers also induced mitochondrial permeability transition, cytochrome c, apoptosis, and cell death. Further, they found mitochondrial depolarization prevented mitochondrial Ca2+ overload, the release of cytochrome c, and cell death. In addition, Sanz-Blasco et al also observed that a series of non-steroidal anti-inflammatory drugs, including salicylate, sulindac sulfide, indomethacin, ibuprofen, and R-flurbiprofen, depolarized mitochondria and inhibited an overload of mitochondrial Ca2+, the release of cytochrome c, and cell death induced by Ab oligomers. Findings from this study suggest that Ab oligomers induce intracellular Ca2+ and promote the entry of Ca2+ into mitochondria, and cause mitochondrial structural and functional damage. Overall, these findings suggest that abnormal calcium homeostasis in relation to mitochondria and Ab oligomers play a key role in AD pathogenesis.
Synaptic damage and the loss of synapses are critical factors responsible for cognitive decline in aged individuals and in patients with AD (Reddy and Beal, 2008). Using electron microscopy, several studies investigated the connection between synaptic loss and cognitive decline in AD (DeKosky et al., 1996; Bertoni-Freddari et al., 1990) and found a 25–30% decrease in synapses in the cortex and a 15–35% decrease in synapses per cortical neuron, suggesting that synaptic loss correlates with cognitive decline in AD patients more than does the number of Aβ plaques, the number of neurofibrillary tangles, or the loss of neurons. Several researchers investigating synaptic proteins in AD patients and healthy control subjects found fewer pre- and post-synaptic proteins in AD patients in comparison to control subjects, suggesting that pre- and post-synaptic proteins are critically involved in AD progression (Gylys et al., 2004; Reddy et al., 2005; Almeida et al., 2005).
Recently, using electron microscopy, a rapid Golgi method, and Bodian staining, Baloyannis et al (2007) investigated synaptic alterations including synapses and dendritic spines in the medial geniculate bodies and the inferior colliculi in subjects with early AD and control subjects. They found substantial neuronal loss and synaptic alterations in the medial geniculate bodies of the AD subjects, as well as in their inferior colliculi. Dendritic spines of the polyhedral and elongated cells of the medial geniculate bodies decreased. Mitochondrial alterations and fragmentation of Golgi apparatus were seen in 15% of the neurons of the medial geniculate bodies and in 5% of the neurons of the inferior colliculi. However, senile plaques and neurofibrillary tangles were not seen in either the medial geniculate bodies or the inferior colliculi. Based on these observations, Baloyannis et al concluded that neuronal loss, and the mitochondrial and synaptic alterations in the medial geniculate bodies and inferior colliculi are involved in the impairment of communication and symbolic sound perception in the early stages of AD.
Findings from electron microscopy, rapid Golgi, Bodian staining, and immunoblotting analyses of brain specimens from AD patients and control subjects suggest that the loss of synapses and synaptic proteins, and alterations in mitochondria and Golgi apparatus are critically involved in synaptic damage and cognitive decline in patients with AD.
Mitochondria are the major source of energy for the normal function of brain cells. Several lines of evidence suggest that Aβ and APP are localized to mitochondrial membranes, interact with mitochondrial proteins, increase ROS production, cause mitochondrial structural and functional damage, and prevent neurons from functioning normally. Further, Ab is transported into mitochondria via the translocase of the outer membrane machinery, and mitochondrial Ab are localized to the cristae of the inner mitochondrial membrane, and Ab are localized to matrix and outer membrane. Oligomeric Ab and g -secretase complex proteins are reported to induce intracellular Ca2+ levels and promote excess accumulation of mitochondrial Ca2+, induce the opening of mitochondrial permeability transition pore, in addition to interaction of mitochondrial proteins and damage mitochondrial intact structure. Based on recent initial APP/Aβ and mitochondrial structural studies, we propose that the overexpression of APP and the increased production of Aβ may cause structural changes in mitochondria, may increase the production of defective mitochondria, and may decrease mitochondrial trafficking and cause abnormal mitochondrial dynamics in neurons that are affected in AD. Further, the numbers of mitochondria per AD-affected neuron (cortical and hippocampal) and per AD-unaffected neuron (Purkinje) need to be quantified in order to determine the connection between the number of healthy and functionally active mitochondria and the synaptic activity in the mitochondria of affected and unaffected neurons in AD mice. The answers to these critical questions may improve our basic understanding of mitochondria, synaptic activity, and cognitive function in AD, and may help define a line of research for the development of treatments to AD patients.
The research for this article was supported by grants from National Institutes of Health (AG028072 and AG026051).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.