Metal for the mind

Neurodegenerative diseases such as Alzheimer’s, Parkinson’s, Huntingdon’s, amyotrophic lateral sclerosis (ALS), and the transmissible spongiform encephalopathies (TSE), or prion diseases (such as scrapie, BSE or Creutzfeld-Jacob disease), have obvious differences in clinical symptoms, but they have a surprising amount in common.

As Claudio Soto, from biotech company Serono International, based in Geneva, Switzerland, pointed out in a recent review, most diseases like this have both sporadic and inherited forms, they all appear later in life, and their pathology is characterised by loss of neurons and damage to the synapses. And they all seem to involve protein misfolding, although the proteins involved are quite different.

In Alzheimer’s it is a 40 or 42 residue peptide, amyloid-?. In Parkinson’s it is ?-synuclein, a 140-residue protein abundantly expressed in presynaptic terminals of vertebrates. In Huntingdon’s disease it is the much larger huntingtin protein. In ALS it is the enzyme superoxide dismutase, a complex of two 151-residue units. In the TSE’s it is the various shorter, prion peptides.

These dissimilar materials form remarkably similar fibrillar deposits, the ’amyloid’ plaques (a fact that has piqued the interest of the polymer science community). But these deposits may not be the cause of the neurodegeneration.

Postmortem studies show a poor correlation between the amount of deposit and the severity of the clinical symptoms, and people without any symptoms of disease have been found to have amyloid plaques in their brains. It has even been suggested that the plaques represent a protective state that sequesters and inactivates toxic oligomers and protofibrils.

Soto proposes two other hypotheses: the misfolded proteins can no longer carry out their intended tasks and this loss of function might damage neurons; or the soluble form of the misfolded protein may itself be toxic (able to disrupt the cellular membranes), or able to induce oxidative stress by catalysing the formation of reactive hydroxyl radicals.

Another common link between these diseases appears to be the involvement of transition metal ions. Many key proteins bind metal ions. The amyloid-? protein binds zinc and copper. The prion proteins bind copper, and have also been found to bind manganese and nickel. Superoxide dismutase binds both copper and zinc, and ?-synuclein binds copper, iron and magnesium.

Observations show that high levels of metals in ground water are found near sites where the diseases seem more prevalent. Clusters of patients with ALS or Parkinson’s disease were found in the Papua New Guinea islands, where the drinking water contains a high percentage of Al³⁺ and Mn²⁺ ions.

Recent investigations of scrapie and Creutzfeldt-Jakob disease in Iceland and Slovakia show that the soil and vegetation there is low in copper and high in manganese ion concentrations. It has even been suggested that the increase in scrapie and BSE cases in the last decade is partially due to acid rain, because acid rain increases the solubility of Al³⁺ and Mn²⁺ ions in subterranean aquifers.

Evidence gathered at the scene of the crime (the results of post-mortem analysis of brain tissue from victims of the neurodegenerative diseases) also appears to implicate transition metal ions. For example, remarkably high concentrations of copper (400?M), zinc (1mM) and iron (1mM) have been found in amyloid deposits in Alzheimer’s affected brains.

But metal ions may not be the villains they appear. Their concentration in the brain is very tightly regulated, and their presence is vital. ’Cells rely on transition metals to regulate a wide range of metabolic and signalling functions,’ comments Henryk Kozlowski, University of Wroclaw, Poland, who is an expert on this subject. ’Their physiological functions are derived from their atomic properties, most notably an incomplete inner valence sub-shell.’ These properties are embodied in the wide range of oxidation states available to transition metal ions, and provide a chemical flexibility that allows them to impose conformational changes upon the proteins to which they bind.

Transition metal ions have good reason to be present at the scene of the crime. In fact, they have good reason to be present in the proteins implicated in neurodegenerative disease. For example, although the in vivo function of the prion protein remains to be confirmed, it has been shown to have superoxide dismutase-like activity when bound to copper.

Superoxide dismutase catalyses the toxic superoxide radicals’ conversion to molecular oxygen and hydrogen peroxide (the active site for this reaction contains one copper and one zinc ion). It is crucial for the cellular antioxidant defence mechanism. Recent data suggest that both the membrane bound amyloid precursor protein, and the amyloid-? peptide implicated in Alzheimer’s disease bind metals because they play a role in metal homeostasis - helping to regulate metal ion concentrations in the brain. Huntingtin protein has been implicated in iron metabolism. And there is evidence that prion proteins facilitate the cellular uptake of copper to carry out perfectly legitimate tasks.

’In the normally folded proteins the metal active site is probably shielded, and access to it is controlled,’ comments Kozlowski. ’However, changes in the conformation of these proteins may expose the active site and make it more prone to act as an oxidative reaction catalyst, or aggregation modulating factor.’ The oxidation reaction he implicates (which may involve either copper or iron) is a Fenton type of reaction where hydrogen peroxide forms a reactive hydroxyl radical. High oxygen consumption, relatively low antioxidant levels, and limited regenerative capacity make neurons particularly susceptible to damage from reactive oxygen species. ’If one looks at amyloid plaques one can see products of oxidation of nucleic acids and proteins that are characteristic of oxygen radical species action,’ comments Kozlowski.

So what is going wrong? What turns Jekyll into Hyde? What causes these proteins and peptides, which appear to bind metals as part of their normal function, to misfold so that their beneficial function is lost, sites for damaging oxidation are revealed, and they aggregate as amyloid fibrils?

As Jeffery Kelly, at Scripps Research Institute, San Diego, US, speaking at a symposium on protein folding at the University of Vermont, US, in September 2004 commented: ’There may be many triggers, from our environment, what we eat, and from our heredity.’ In the case of Alzheimer’s, Kelly has recently identified a possible role for cholesterol as the villain.

Oxidised cholesterol metabolites contain an aldehyde group, which has been shown to condense with the amines in the amyloid-? peptide. The presence of this large covalently attached hydrophobic group considerably reduces the complex’s solubility and encourages it to misfold. In a nice analogy with an industrial production plant, more faulty or marginal products (proteins that are more prone to misfolding) are released when the QA department is overworked and understaffed. ’Failures of quality assurance in the endoplasmic reticulum allow faulty proteins, already on the edge of misfolding, to be released,’ commented Kelly.

In the case of Huntington’s disease, the finger of blame clearly points to genetics. Sufferers have an excess number of glutamine groups on one particular segment of the huntingtin protein, and this causes the misfolding.

A widely accepted cause of prion diseases is an already misfolded peptide templating the misfolding of others. In this model the misfolded peptide becomes an agent-provocateur, creating disease in any new system to which it is introduced. But several alternative, or additional, models have been proposed, some involving the transition metal ions.

In vitro experiments carried out by David Brown, Bath University, UK, have shown that when manganese replaces copper in the prion protein, the secondary structure is changed. The prion protein becomes less stable, and quickly converts to the misfolded form. This may explain the high prevalence of scrapie in areas of Iceland where the vegetation contains up to 1000 times the levels of manganese in unaffected areas. It could be that the prion protein itself encourages the uptake of manganese into the cells.

Ashley Bush from Harvard medical school, US, proposes a metal-ion related mechanism for Alzheimer’s. His research has shown that when metal ion concentrations are high the amyloid-? protein changes shape to bind up to 3.5 moles of copper or zinc with varying affinities. These ’hypermetallated’ complexes can be oxidised releasing soluble forms of the peptide, rogue enzymes that resist clearance. In a healthy individual there is very little soluble amyloid-? protein in areas of the brain such as the cortical synapse. But in an individual with Alzheimer’s the naturally high Zn²⁺ concentrations in the synapse cause increasingly prevalent protein to precipitate, creating a reservoir of potentially toxic material.

Both these models imply a failure in metal homeostasis - the physiological processes that allow the body to maintain equilibrium. Tom O’Halloran, from Northwestern University, US, suggests such failures may be triggers for what he refers to as the ’metal trafficking diseases’ in certain groups of people. ’There are quite robust systems in place to control metabolism of essential metals such as iron, copper, zinc and manganese,’ he comments. ’However, genetic variations in genes controlling metal uptake, utilisation or efflux may predispose some individuals to these diseases. In those cases the individual may be more susceptible to elevated levels of essential metals from environmental sources such as drinking water.’

The control mechanisms may also break down with age - explaining these diseases’ late onset. O’Halloran comments that non-essential metals, such as lead and mercury, are an entirely different matter: these may be able to piggyback into the cell via essential metal transport systems and do significant harm to organs such as the brain. There can be several different manifestations of this metal ion poisoning - including the onset of neurodegenerative disease.

Study of the neurodegenerative diseases is proving a fascinating and revealing area. We are gaining fundamental insight into the role of protein folding and the critical importance of metal ions in biological processes. But is this taking us further towards preventative therapies, or a cure?

Since it could be reactive oxygen species, generated by reactions involving the metal centres of misfolded proteins, that do the actual damage, one approach might be to increase the intake of antioxidants. Epidemiological studies have shown a moderate risk reduction for Parkinson’s disease in green tea drinkers, improved cognitive performance for Alzheimer’s patients who consumed Ginkgo biloba leaves, and a decreased incidence of Alzheimer’s for wine drinkers in France.

Back at Harvard, Bush has proposed two possible therapeutic options. Since his model implicates excessive, pathological, metal binding, he and his collaborators hope to identify materials that selectively block metal binding sites on peptides such as amyloid-?.

However, he comments that without any metal ions the residues at the binding sites are relatively unstructured, and as a result designing classic ’lock and key’ inhibitors is problematic. Another possible approach is to identify molecules that will effectively compete with the peptide for the metal ions. Epidemiological studies have shown a protective action for drugs that are known chelators (for example ibuprofen).

Many other approaches are being investigated. Scientists from Novartis, a leader in neuroscience for more than 50 years, recently published a list of the drugs in research and development for neurodegenerative diseases, and the intended modes of action for each one. They wanted to give the public an idea of the complexity of drug discovery in this &0xFB01;eld. There were 551 drugs on the list, under consideration at more than 200 pharmaceutical companies and universities.

’There are many scientific approaches to therapies such as aggregation inhibitors, inhibition of processing enzymes such as the secretases for Alzheimer’s, removal of toxic moieties such as the amyloid-? peptide using antibodies, and anti-inflammatory agents such as NSAIDs,’ comments Graeme Bilbe, global head for the neuroscience disease area, at the Novartis Institutes for BioMedical Research in Basel, Switzerland. This significant level of effort, both in fundamental research and drug development is appropriate because, as Bilbe notes, ’neurodegenerative diseases are an increasingly important issue in our ageing society.’

Fiona Case is a freelance science writer based in Vermont, US