Oxidation, Mitochondrial damage and Cytotoxicity in Alzheimer’s Disease
Introduction to Alzheimer’s
In 1907, Dr Alois Alzheimer first characterised a disease of memory loss and dementia in elderly patients. It is the most common cause of senile dementia, affecting almost 40% of individuals over 85 years of age. The majority of cases are apparently spontaneous, although a small number of gene mutations are associated with increased risk and earlier onset of disease in some families. The disease involves extensive cortical atrophy, visible on brain sections as abnormally wide sulci.
An explanation of the cause of cell death has long been elusive. Numerous studies have been published that show cell death to be entirely by apoptosis, while another set have shown exclusively necrosis. This obviously suggests that either outcome can occur depending on the situation. What is almost totally certain is the involvement of the plaques of crystalline protein found extensively in Alzheimer’s patient brains: amyloid.
A complex picture is emerging of how this amyloid plaque is associated with cell death, with the precise mechanisms becoming increasingly clear.
A likely culprit: The Amyloid-beta peptide
The plaques found in the brains of Alzheimer’s disease sufferers are composed of amyloid fibrils, a material made up of an ordered crystalline precipitate of misfolded protein. Amyloid occurs when some form of environmental stress disrupts the folding pattern of a normally soluble protein. The disrupted form, either broken into fragments or seriously contorted, is no longer soluble in solution. It converts into flat beta-sheets and forms ordered stacks which accumulate into long protofilaments. These twist together like strands in a rope to form fibrils, which aggregate into large plaques such as those seen around brain cells in Alzheimer’s disease.
The protein responsible for Alzheimer’s amyloid is APP, or Alzheimer’s Precursor Protein, is a 770 residue protein whose precise function is unclear. APP can be degraded by proteases along two pathways, only one of which generates the 49-43 residue A-Beta peptides. Which pathway is taken appears to be related to cholesterol levels.
These A-Beta peptides, henceforth referred to as A-Beta, are capable of generating reactive oxygen species, or ROS, through a series of redox reactions with metal ions. They have been shown to do this in vitro and in the absence of cellular material of any kind. As ROS can cause serious damage to cells, this ability immediately suggests a mechanism for cytotoxicity.
It appears increasingly likely that the fully formed plaque itself is not the cause of the disease. Plaque is found in the brains of many elderly people who have retained entirely normal brain function, while some patients with severe Alzheimer’s have only minimal levels. This alone cannot discount the idea that A-Beta is responsible, however, when there is extensive biochemical evidence suggesting its involvement in several different mechanisms of cell damage. While the fully formed and relatively stable amyloid fibrils may not be cytotoxic, the smaller precursor elements are probably responsible. While monomeric A-Beta is not considered neurotoxic, small diffusible oligomers have been shown to be neurotoxic in mice.
The importance of oxidation
The simple explanation for the majority of cell death in Alzheimer’s disease is oxidative damage. Autopsies of Alzheimer’s sufferers show extensive oxidative degradation of proteins and DNA present in areas of the brain affected by the disease. This evidence is backed up by studies which found the progression of neurodegeneration was impeded by the addition of high levels of antioxidants.
Oxidative damage would also explain one of the first problems observed in cell cultures when A-Beta is added: rapid mitochondrial problems, usually involving the failure of proton pumping. Such mitochondrial failure is a common symptom of oxidative stress on a cell: mitochondrial enzymes are extremely sensitive to oxidative damage. Worse still, disruption to mitochondrial function can lead to the production of more ROS inside the organelle, exacerbating the damage.
Damage to mitochondrial membrane integrity could very easily kill a cell. If the outer membrane is perforated and Cytochrome c escapes into the cytoplasm, the pathway of internal activation of apoptosis via caspase activation will be triggered. Apoptosis could also possibly be triggered if serious oxidative damage to DNA occurs.
There are also more brute-force ways that attack by ROS can prove lethal to a cell. Disruption of membrane stability, structural proteins or normal enzyme activity could prove sufficiently traumatic to a cell to result in necrosis. Whichever pathway to cell death is involved in any particular case, the importance of oxidative damage is well established. One question remains: how could the A-Beta peptide cause these effects?
The production of radicals
As the importance of oxidation and the creation of ROS has been evident for some time, much thought has gone into investigating the way A-Beta could be responsible. The first and most obvious explanation involves the ability of A-Beta to react with copper and produce reactive species.
A-Beta and copper redox reactions
A-Beta’s ability to insert in bilipid layers ensures the presence of reactive oligomers just outside the cell. There, in close association with the membrane, they may undergo a series of redox reactions with trace levels of copper present in the brain, generating ROS. The close proximity to the membrane means that the most likely result of ROS attack will be lipid peroxidation, damage to channels and transporters and ultimately serious disruption of ion balance. Such disruption puts the cell under significant oxidative stress, with all the associated mitochondrial failure which is an early sign of Aâ toxicity.
Despite the elegance of this model, the precise mechanisms involved were not at all well explained. It now appears that despite its logical merits, experimental evidence indicates that an alternative mechanism is being used.
The role of direct perforation and Calcium influx
Recent studies have indicated that the presence of A-Beta in the membrane alone is not directly related to increases of ROS inside the cell. Two interesting factors were observed: firstly, that ROS generation was dependent on rising intracellular calcium. Secondly, and oddly enough, that rises in ROS are not seen at all in neurons, while they are seen in astrocytes. This suggests that some unique characteristic of astrocytes is responsible for the lion’s share of ROS generation.
The apparent relationship between calcium and ROS depends on the ability of A-Beta amyloid to insert into the membrane and perforate it. This provides an elegant explanation for why single monomers A-Beta are not toxic, but tetramers are: the transmembrane domains of the tetramers can combine to form a pore. This would allow an influx of extracellular calcium. Disruption of the normal homeostatic balance of the cell would result, putting it under a great deal of stress and promoting oxidation.
The relationship between this calcium influx and generation of ROS in astrocytes is explained by the discovery that these cells contain an enzyme previously only identified in immune phagocytes. NADPH oxidase is used by these cells to produce large quantities of ROS for use as a weapon against microbes. It is found associated with mitochondria, and is activated by increased intracellular calcium levels.
Alternatives to oxidative damage: An enzyme binding model
One flaw in the membrane perforation theories is that there is evidence that it does not occur at all in neurons. This, added to the fact that neurons lack NADPH oxidase and thus cannot undergo the mechanism of ROS generation outlined above, raises some interesting questions. It is well established that in Alzheimer’s disease there is extensive neuronal death, and also that A-Beta fibril intermediates are neurotoxic. Are neurons dying because of lack of astrocyte support, or because of the toxic debris released by nearby mass necrosis? The neurotoxicity of A-Beta in pure cultures of neurons suggests otherwise. Is A-Beta causing an alternative mechanism of cell death in neurons?
What is A-Beta AD and how could it be involved?
Yeast-2 hybrid studies of whole-brain homogenates have yielded only one protein which binds to A-Beta peptide, which has subsequently been named A-Beta AD: Amyloid Beta binding Alcohol Dehydrogenase. It has been identified as a mitochondrial enzyme, and knockout models have shown it to be vital for survival in Drosophila. Most interesting is its binding characteristics: once bound to the amyloid fragments, the structure of A-Beta AD was shown to be highly abnormal. The enzyme binding site had clearly been completely disrupted.
Could it be that the presence of amyloid fragments causes mitochondrial toxicity by binding with and deforming a critical enzyme?
An apparently gaping flaw in this argument is the fact that A-Beta is inherently hydrophobic, as reflected by its drive to form amyloid fibrils to protect itself from the surrounding aqueous environment. Why would it leave the relative safety of the membrane and enter the cytoplasm? How and why it does so remains unclear, but there is very convincing evidence that it can be found throughout neurons, particularly in the mitochondria. Recent studies by confocal microscopy show plenty of A-Beta can be found in the mitochondria, along with A-Beta AD. This was supported by immunogold electron microscopy images.
Conclusions
The ability of the A-Beta peptide to cause cell loss in the brain is clearly more complex than was understood a few years ago. Specific mechanisms are now becoming known, providing more possible targets for therapy and prevention. Perhaps most importantly, it is now evident that the fully formed plaque is not toxic and attacking it will not be of clinical use. The reactivity of individual A-Beta peptides is another culprit which has been largely exonerated as the result of recent discoveries. It is the pore-forming, enzyme-binding behaviour of oligomers that needs to be the focus of future research.
An excellent review of this area is by Laura Canevari et al, in Neurochemical Research, Volume 29, pages 637-650: 2004. The study on ABAD colocalisation was by Joyce Lustbader et al, in Science, volume 304, pages 448-452, 2004. This has been a Node your Homework production.