4. Parkinson's Disease

Parkinson's disease (PD) was initially described as Shaking Palsy in 1817 by the English surgeon James Parkinson. The second most common form of motor system degeneration, it is characterised by a progressive loss of dopaminergic neurons in the substantia nigra pars compacta in the ventral midbrain.

There is strong evidence implicating the abnormal processing of a number of cellular proteins via the ubiquitin/26S proteosomal system in PD. Such proteins are present in the cytosol as insoluble, unfolded, ubiquitinated polypeptides. The characteristic inclusion bodies of PD, Lewy bodies, (LB) present in affected neurons in the brains, are eosinophilic intracytoplasmic inclusions, with a core of granular and filamentous material surrounded by radiating filaments 10-15nm in diameter, Figure 4. Wild type α-synuclein is found as a major component of Lewy bodies, suggestive that α-synuclein is involved in the pathophysiology of PD and are present within dopaminergic neurons, axons and synapses of the substania nigra.

Synucleins are a family of small (between 123-143 amino acids), highly charged proteins, expressed predominantly in neural tissue and in some tumours. Three human synuclein proteins α, β and γ are known, and they are encoded by separate genes on three different chromosomes, 4q21.3-q22, 5q23 and 10q23.2-q23.3. The physiological function of synucleins is currently unclear, although some data suggests a role in the regulation of membrane stability and/or turnover. It has also been proposed that α-synuclein may play an important biological role either in the modulation of neurotransmitter release, as a regulatory protein that can bind and inhibit tyrosine hydroxylase, the rate-limiting enzyme in dopamine biosynthesis or possibly in synaptic organisation. The family, known as the Iowan kindred described recently exhibit a triplication of the α-synuclein gene, which results in autosomal dominant PD giving four copies of the gene rather than the usual two. This clearly implies that overproduction of wild type α-synuclein may be the cause of PD rather than simply the production of mutant protein.

The presence of Lewy bodies in affected regions of PD brains indicates improper handling of this protein. Alpha-synuclein may be the actual building block of the fibrillary components of LBs, binding tubulin amongst a host of other ubiquitinated proteins, synphilin-1 and parkin and acts as the focal point for the aggregations. Its polymerisation is associated with concomitant changes in secondary structure which can range from the natively unfolded state in solution, to α-helical (random coil) in the presence of lipid containing vesicles to anti-parallel β- pleated sheet structures or amyloid structure in fibrils. Such products will initiate disturbances in the cytosolic cellular compartment interacting with vesicles, dopamine transporters and intraneuronal mitochondria, which may lead to cellular death..

A link between iron and α-synuclein has been identified in vitro. Iron enhances intracellular aggregation of α-synuclein leading to the formation of advanced glycation end products while α-synuclein liberated hydroxy radicals when incubated with Fe⁺⁺Extracellular α-synuclein has recently been found to generate ROS. Pre-treatment of cells with cell-permeable iron chelators, transferrin receptor antibodies or transfection with glutathione peroxidase inhibited intracellular oxidant generation, α- synuclein expression/aggregation as well as apoptosis. Interestingly magnesium inhibited aggregation, by preventing conformational changes.

Parkin is located on chromosome 6, 6q25-27, and has a distinct domain structure (Figure 5) consisting of a ubiquitin-like (Ubl) domain at its amino terminus and two carboxy-terminal cysteine-rich RING[1] (really interesting new gene) fingers flanking an in-between RING (IBR) domain (Cookson, 2003a). Wild type Parkin has been shown to have E3 ubiquitin-protein ligase activity, whose cellular role is to add the polyubiquitin tail to targeted proteins for their degradations, through the binding of their Ubl to proteasomes.

Since the cysteine residues of Parkin are integral to its E3 activity, their modification by reactive oxygen species, (possibly catalysed by the increasing iron accumulation), may impair its function. Two recent reports indicate that S-nitrosylation may alter Parkin's ability to ubiquitate target proteins, although they report opposite effects. Peroxide will induce mis-folded Parkin which can be prevented if heat shock chaperones, Hsp70 and its co-chaperone Hsp40, are induced. The three carboxy-terminal amino acids of Parkin are necessary for proper folding as well as function. Impairment of proteosomal activity by stable expression of parkin mutants leads to accumulation of oxidised proteins and lipids, thereby sensitising the neurons to various forms of stress leading to neuronal death. Caspase 1 and 8 can directly cleave Parkin leading to loss of ubiquitin ligase activity, causing accumulation of toxic parkin substrates and triggering dopaminergic cell death. Lymphocytes from patients with AR-JP, homozygote for the Cys212Tyr parkin mutation, show increased sensitivity to dopamine, iron and hydrogen peroxide.

Synphilin-1 is an alpha-synuclein-binding protein that is ubiquitinated by Parkin, and promotes the formation of cytosolic inclusions. It also interacts with the E3 ubiquitin-ligases SIAH-1 and SIAH-2 (Liani et al., 2004). SIAH proteins ubiquinate synphilin-1 in vivo, and promote its degradation by the ubiquitin-proteasome system. An inability of the proteasome to degrade the synphilin-1/SIAH complex will lead to the formation of ubiquitylated cytosolic inclusions. The importance of this protein in Parkinson's disease awaits further clarification although SIAH immuno-reactivity has been shown in Lewy bodies.

Other proteins, Park 3, Park 4, Park 5 (UCHL-1), Park 6 (PINK1), Park 7 (DJ-1) Park 8-11 and NR4A2 may play an important role in Parkinson's disease but further clarification at the protein levels is required.

An inability to release ubiquitin from the polyubiquitin tail, as a result of reduced UCH L1 activity, will lead to incomplete degradation of the target protein and accumulation of neurotoxic proteins, Figure 6. The protein level of UCH L1 is down regulated in the brains of idiopathic Parkinson's disease. UCH L1 protein is a major target for oxidative damage, with carbonyl formation, methionine and cysteine oxidation.

An inevitable consequence of ageing is an elevation of brain iron; e.g. in the putamen, motor cortex, prefrontal cortex, sensory cortex and thalamus, although its localisation into various iron storage proteins remains controversial. The concentrations of iron vary considerably between different brain regions; the only differences in the brains of Parkinson's patients is that there is a specific elevation of iron in the substania nigra and the lateral globus pallidus, by approximately 2 fold in comparison to age-matched controls (Götz et al.,2004). The cause of such changes in brain iron content in specific brain region remains undefined, and may be caused by a number of factors which include changes in iron release mechanisms across the BBB; or the regulation of iron transport across the membranes of specific brain regions. There is increased levels of iron in both Lewy bodies and within cytosolic compartments of dopminergic neurons of the substantia nigra in PD patients which will cause oxidative damage.

Cumulative evidence supports an 'oxidative stress hypothesis' for initiation of nigral dopamine neuron loss, Figure 7. Substantia nigra has a relatively high metabolic rate, with a high content of dopamine, neuromelanin, polyunsaturated fatty acids and iron but low antioxidant protection, e.g. glutathione. Therefore the imposition of oxidative stress will readily overwhelm the cytoprotective and antioxidant capacity, leading to the initiation, propagation and programmed cell death (apoptosis) of the dopaminergic neurons. It is remarkable that, in contrast to other iron storage diseases, where 10-20 fold iron increases in iron stores must be attained before clinical abnormalities occur, e.g. untreated genetic haemochromatosis, GH, and thalassaemia patients, THAL, it requires only a two fold increase in the iron content of the substantia nigra of PD brains to produce extensive pathological consequences. Although haemosiderin is the predominate iron storage protein in GH and THAL, it is of interest that in PD ferritin and neuromelanin are the predominant iron storage protein; Ferritin increases in the microglial cells which are in close proximity to the degenerating neurons of the SN of PD, (reviewed in Götz et al., 2004) while H ferritin is reported to be the predominant form, (L ferritin decreases) in substantia nigra and lateral globus pallidus of PD brains by comparison to that of age matched controls.

Homozygous knock-out mice for H ferritin die in utero, while heterogenous mice are viable and show brain iron comparable to controls although H-ferritin content is half that of controls. However transferrin, transferrin receptors, L-ferritin, DMT1 and caeruloplasmin were increased. Oxidative stress was increased in these mice, as exemplified by oxidatively modified proteins and reduced activities of cytoprotective enzyme, e.g. superoxide dismutase .

Cells which over-expressed human H-chain ferritin accumulated the protein at levels 14-16-fold over background, and show an iron-deficient phenotype manifested by a 5-fold increase of iron responsive protein activity, 2-2.5-fold increases in both transferrin receptor, and iron-transferrin iron uptake, and approximately 50% reduction of the labile iron pool. Over-expression of the H-ferritin strongly reduced cell growth and increased resistance to H₂O₂toxicity; these effects were reverted by prolonged incubation in iron-supplemented medium.

It is noteworthy that despite the increasing brain iron content in PD there is no reciprocal up-regulation of L-ferritin expression in response to the iron increase (Dexter et al., 1991). Iron regulatory proteins, IRP-1 and IRP-2 act as iron sensors and regulate ferritin synthesis. Studies have therefore been undertaken to ascertain whether changes in post transcriptional control of ferritin synthesis have occurred in SN of Parkinson's patients. Since changes in ubiquination appear to be an important facet of Parkinson's disease, changes in the degradation of IRP-2 by ubiquitination, may be an explanation for the high cellular iron content within the substantia nigra and the lateral globus pallidus. No studies of IRP-2 binding activity have been reported, although studies of IRP-1 in the SN of Parkinson's patients showed no significant changes by comparison to age matched controls. This may explain why ferritin mRNA translation is not upregulated despite the increased iron accumulation. However recent evidence has suggested that IRP2 dominates post-transcriptional regulation of iron metabolism in brain such that further studies are currently needed to clarify the roles played by IRP-1 and IRP-2 in controlling iron homeostasis.

No IRP-2 polymorphisms are reported in subjects with sporadic PD and normal controls which might have played an important role in the development of the disease. Genetically engineered mice which lack IRP-2 but have the normal complement of IRP-1, develop adult onset neurodegenerative disease associated with inappropriately high expression of ferritin in degenerating neurons while mice that are homozygous for a targeted deletion of IRP-2 and heterozygous for a targeted deletion of IRP-1 developed severe neurodegeneration with severe axonopathy, with increased levels of ferric iron and ferritin expression as well as neuronal cell bodies degenerating in the substantia nigra.

In milder forms of Parkinson's disease, no change in SN iron content is reported, perhaps suggesting, that the iron deposition might be related to the pathogenesis rather than to the etiology of the disease. The increase in SN iron content, detectable in 90% of individuals affected by the disease by ultrasound measurements, was also detectable in 45% of relatives of Parkinson's disease patients, indicating a degree of inheritance of this disorder.

Large amounts of iron are sequestered in substantia nigra and in locus coerulus as a neuromelanin-iron complex in dopaminergic neurons (Zecca et al., 2003) particularly in Parkinson's Disease.Neuromelanin, a granular dark brown pigment, is produced in catecholaminergic neurons of the SN and locus coeruleus and is possibly the product of reactions between oxidised catechols with a variety of nucleophiles, including thiols from glutathione and proteins (Götz et al., 2004). The function of neuromelanin in the pigmented neurons is unknown but it could play a protective role via attenuation of free radical damage by binding transition metals, particularly iron. In normal individuals, the neuromelanin-iron complex is found in both the substantia nigra and locus coeruleus and increases linearly with age in the substantia nigra. Whether the ability of the neurones to synthesis neuromelanin is impaired in PD patients is unknown, although it is reported that the absolute concentration of nigral neuromelanin is less than 50% in PD with respect to age matched controls. In vitro it has been shown that melanin can bind a significant amount of iron at two sites although the pigment appears to be only 50% saturated with iron in PD. Iron is bound to the catechol groups. EPR studies of the SN show that ferric iron is bound to neuromelanin as a high spin complex with an octahedral configuration. Mossbauer spectroscopy, MS, shows that the ferric iron is bound in ferritin-like oxyhydroxide clusters, the spectra obtained were comparable to that of ferritin.