Etiology and Pathogenesis of Parkinson’s Disease

Parkinson's disease (PD) is the second most common neurodegenerative disease after Alzheimer's disease. It is urgently needed to elucidate the cause of the disease and to establish neuroprotective treatment. We have been working on the etiology and pathogenesis of PD for many years and we found selective loss of mitochondrial complex I and the alpha-ketoglutarate dehydrogenase complex in the nigral neurons of patients with PD. Our observation firmly established mitochondrial defects in PD. Mitochondrial respiratory failure induces oxidative damage in neurons, and we found increase in hydroxynonenal and 8-oxo-deoxyguanine, indices of oxidative damage, in the nigral neurons of PD. These abnormalities can trigger apoptotic cell death. The primary events which induce mitochondrial failure and oxidative damage are not known, however, it has been postulated that the interaction of genetic risk factors and environmental factors would initiate the degenerative process. Based on this assumption, we conducted genetic association studies by the candidate gene methods. We found that polymorphic mutations of superoxide dismutase-2 and 24-kDa subunit of mitochondrial complex I were associated increased risk of developing Parkinson's disease. While we were doing this genetic association study, we found a family, in which parkinsonian phenotype completely segregated with a polymorphic mutation of the superoxide dismutase-2 gene. In this family, 4 out of 6 siblings were affected with early onset parkinsonism and the parents were apparently normal. Thus the mode of inheritance appeared to be autosomal recessive and this type is now called as AR-JP or Park2. We confirmed the linkage of this type of familial Parkinson's disease to the superoxide dismutase loci that is located in the telomeric region of chromosome 6 by the linkage analysis using microsatellite markers in this region. Then we found another family, in which an affected patient showed lack of one of the microsatellite markers (D6S315), which we were using in the linkage analysis. This observation prompted us to initiate the molecular cloning of the disease gene utilizing D6S315 as the initial probe. The molecular cloning was done with the collaboration with Professor Nobuyoshi Shimizu of Keio University. We identified a novel gene and confirmed that mutations of this novel gene were found only in the patients with autosomal recessive Parkinson's disease. The novel gene was named parkin. We conducted mutational analysis on more than 700 families with Parkinson's disease. We also established a method to detect compound heterozygotes of parkin mutations. Mutinous of the parkin gene were found in approximately 50% of autosomal recessive families. Many kinds of exonic deletions and point mutations were found. This type of familial Parkinson's disease had been considered to be unique among Japanese, but since we started mutational analysis of the parkin gene, we confirmed the world wide distribution of parkin gene mutations. Then we analyzed functions of parkin protein with the collaboration with Dr. Keiji Tanaka of Tokyo Metropolitan Institute of Medical Sciences. We found that parkin protein was a ubiquitin-protein ligase of the ubiquitin system. Now we are working on the candidate substrates of parkin protein as a ubiquitin ligase. We found that CDCrel-1, a synaptic vesicle protein, was a candidate substrate of parkin protein. In addition, we found two additional candidate proteins, i.e., alpha-synuclein 22 and PAEL receptor, with the collaboration of Professor Denis Selkoe of Harvard Medical School and Dr. Ryosuke Takahashi of RIKEN, respectively. Accumulation of PAEL receptor in the endoplasmic reticulum causes endoplasmic reticulum stress and apoptotic cell death. We found evidence to indicate accumulation of PAEL receptor and the presence of endoplasmic reticulum stress in a patient with AR-JP (Park2). Thus our studies firmly established that a genetic defect of an enzyme in the ubiquitin-proteasome system induces selective nigral neuronal death. We indicated the important role of the ubiquitin-proteasome system in neurodegeneration in general. In many other neurodegenerative disorders, such as Alzheimer's disease, Huntington's disease, Machado-Joseph disease, dentatorubral-pallidoluysian atrophy, and ALS, ubiquitinated proteins are accumulated in neurons. Thus protein handling in the ubiquitin-proteasome system appears to be affected in these neurodegenerative disorders despite the difference in the primary defects. Our studies also suggest many potential approaches for the discovery of neuroprotective treatment for not only Parkinson's disease but also other neurodegenerative disorders.

On the other hand, MPTP or a combination of maneb and paraquat are environmental risk factors for PD. Several PD-related genetic factors (e.g., SNPs insynuclein) have been discovered, and the interaction between genetic and environmental factors might contribute to sporadic PD.

Environmental factors and PD
2.1 1-methyl-4phenyl-1,2,3,6-tetrahydro-pyridine (MPTP) Langston et al reported on four patients who developed typical parkinsonism after repeated injection of MPTP intravenously (Langston & Ballard 1983). Furthermore, these patients responded to treatment with levodopa and bromocriptine and experienced motor fluctuations, similar to patients with idiopathic PD. Interestingly, fluorodopa positron emission tomography (FD-PET) study of subjects exposed to MPTP demonstrated that transient exposure to this toxin could lead to progressive nigral degeneration (Vingerhoets et al. 1994). Neuropathological findings of 3 subjects with MPTP-induced parkinsonism showed moderate to severe dopaminergic neuronal degeneration without Lewy bodies in the substantia nigra. Interestingly, there was gliosis; clustering microglia around nerve cells and extraneural melanin. These findings indicated active and continuing neuronal cell degeneration (Langston et al. 1999). PET and neuropathological studies suggested that an active neuronal cell death process was persistent for years after these subjects were exposed to MPTP. MPTP is a by-product of the chemical synthesis of a narcotic meperidine analog. This chemical agent is highly lypophilic, therefore it quickly crosses the blood-brain barrier and converts to 1-methyl-4-phenyl-2,3-dihydropyridinium ion (MPP+), which is the active toxic compound, via monoamine oxidase B within nondopaminergic cells, such as glial cells and serotonergic neurons. MPP+ is transported into dopamine neurons by the dopamine 3 transporter, and therefore exhibits selective toxicity to dopaminergic neurons. Furthermore, MPP+ accumulates in mitochondria, where it inhibits the mitochondrial electron transport chain component complex I (Przedborski & Vila 2003). Mizuno et al reported that MPP+ inhibited the -ketoglutarate dehydrogenase complex of the mitochondrial tricarboxylic acid cycle, which synthesizes succinate from α−ketoglutarate (Mizuno et al. 1987). In SNc of sporadic PD subjects, decreased activity and protein levels of complex I have also been reported (Schapira et al. 1989). Thus biochemical changes in dopaminergic neuronal degeneration in sporadic PD were essentially similar to those of subjects exposed to MPTP.

Pesticides
Epidemiological studies have suggested the association between PD and exposure of pesticides such as rotenone and paraquat. Rotenone is a potent and high-affinity specific inhibitor of complex I. In rodents, continuous infusion of these toxic agents induced dopaminergic neuronal cell death and inclusion bodies, which were similar to Lewy bodies in PD patients (Betarbet et al. 2000). Therefore, both MPTP and rotenone disrupt mitochondrial function and play important roles in the nigral cell degeneration in PD. Paraquat, which is a member of a chemical class known as bipyridyl derivatives, is one of the most commonly used pesticides, and leads to oxidative and nitrative stress (Berry et al.). The chemical structure of paraquat is similar to MPP+. Epidemiologic studies implicate the association between exposure to this toxin and the development of PD. Interestingly, recent studies revealed that exposure to a combination of paraquat and maneb, which is also widely used as an agricultural pesticide, exacerbate dopaminergic degeneration in the rodent model and cause an increased incidence of PD in humans (Costello et al. 2009). In this context, exposure to a combination of several toxins could lead to greater risk of developing PD than a single toxin.

Caffeine, cigarette
Smokers and coffee drinkers have been associated with a lower risk of PD in several epidemiological studies. Hernan et al. conducted a meta-analytic review for the association between caffeine intake, smoking, and the risk of PD. Interestingly the cigarette smokers had 60% lower risk for the development of PD than people without a history of smoking (Hernan et al. 2002). Some experiments suggested that chronic nicotine treatment could have protective effects against the nigrostriatal neuronal cell death in MPTP-treated animal models. Several studies revealed that cigarette smoking inhibited monoamine oxidase (MAO) activity and nicotine stimulated dopamine release. As a result, cigarette smoking may suppress free radical generation via MAO-B-associated metabolism of dopamine, and be able to protect against dopaminergic cell death (Miller et al. 2007). Similar to cigarette smoking, coffee drinking could result in a 30% decreased risk of PD compared with non-coffee drinkers (Hernan et al. 2002). Chen et al reported that caffeine had similar effects as A 2A antagonists, and attenuated MPTP toxicity via A 2A receptor blockade ).

Disease process and dual hit theory
Braak and colleagues proposed neuropathological staging in PD. They reported that the dorsal motor nucleus of the glossopharyngeal and vagal nerves and the anterior olfactory nucleus were initially affected, and subsequently the substantia nigra and locus coeruleus became involved. Cortical areas showed less vulnerability, gradually becoming affected (Braak et al. 2003). These pathological findings suggest that the neuronal degeneration in PD may extend from peripheral systems, such as olfactory and autonomic systems to cortices. They also proposed the so-called "Dual-Hit" hypothesis, which suggests that neurotropic pathogens may enter the nervous system via both nasal and intestinal epithelium (Hawkes et al. 2007). These events may promote α-synuclein aggregation and subsequently misfolded α-synuclein may propagate between cells in a prion-like manner (Angot et al. 2010). By using cultured cell models and α-synuclein transgenic mice models, Desplats et al demonstrated direct neuron to neuron transmission of α-synuclein aggregates (Desplats et al. 2009). Actually, in the brains of PD patients who received fetal ventral mesencephalic transplants, grafted nigral neurons contained Lewy body-like inclusions , Kordower et al. 2008. These results corroborate the hypothesis that progression of αsynuclein aggregation relies on the same mechanisms of prion propagation. In conclusion, environmental factors could play important roles in the pathomechanisms of neuronal degeneration in sporadic PD

α-synuclein
-synuclein is a 140-amino acid protein that is abundantly expressed throughout the brain, and especially in presynaptic nerve terminals. In 1997, an -synuclein A53T mutation was isolated from an autosomal dominant PD case. Cases of sporadic PD and A53T-associated patients displayed similar clinical phenotypes, with the exception of a relatively earlier age of onset and atypical features such as cognitive decline, severe central hypoventilation, orthostatic hypotension, myoclonus and urinary incontinence (Polymeropoulos et al. 1997, Spira et al. 2001). In addition, A30P (Kruger et al. 1998) and E46K (Zarranz et al. 2004) mutations were also identified. Following the identification of mutations in the -synuclein gene, multiplication of SNCA has also been identified to cause autosomal dominant familial PD (Ibanez et al. 2004, Chartier-Harlin et al. 2004, Nishioka et al. 2006, Ahn et al. 2008. Considering that -synuclein protein is the major component of Lewy bodies (Spillantini et al. 1997), aggregation of -synuclein is thought to be a key event in dopaminergic neuronal cell death in both SNCA-linked and sporadic PD (Lee & Trojanowski 2006). -synuclein has a high propensity to aggregate, just like other proteins associated with neurodegenerative disease, such as tau, amyloid precursor protein, and poly Q proteins (Bossy-Wetzel et al. 2004, Cookson 2005). -synuclein exists in solution as an unstructured monomer, however, monomeric -synuclein can form insoluble fibrillar aggregates with an antiparallel -sheet structure through the formation of oligomeric forms (protofibrils). The fibrils assemble in vitro, and these filaments closely aggregate just like pathologic inclusions in PD or MSA (Cookson 2005). Fibril formation by pathogenic mutations is accelerated in vitro compared with the wild type (Narhi et al. 1999, Conway et al. 2000, Greenbaum et al. 2005. In several experimental models, including cultured cells, rodent, Drosophilae and Caenorhabditis elegans, pathogenic mutants also promote self-aggregation and oligomerization into protofibrils, compared with the wild-type protein. Some studies suggest that -synuclein could participate in a protein degradation pathway such as the ubiquitin-proteasomal system (UPS) (Rideout et al. 2001, Zhang et al. 2008, and chaperone-mediated autophagy (Cuervo et al. 2004).
The functions of -synuclein under normal physiological conditions remain unknown. However, several studies have shown that -synuclein associates with synaptic vesicles and modulates neurotransmitter release (Jensen et al. 1998, Li et al. 2004. Electrophysiological study in -synuclein knockout mice revealed that -synuclein could act as an activitynegative regulator of DA neurotransmission at synaptic terminals (Abeliovich et al. 2000).synuclein is also speculated to have various other functions, such as modulation of tyrosine hydroxylase (TH) activation (Perez et al. 2002), inhibition of vesicular monoamine transporter-2 (VMAT2) activity , and interaction with septin4, which have been implicated in exocytosis (Ihara et al. 2007). Recently, Cooper and colleagues reported that -synuclein blocked endoplasmic reticulum (ER)-to-Golgi vesicular trafficking, and induced ER stress followed by cell death (Cooper et al. 2006). In addition, -synuclein may interact with the synaptic membrane through lipid rafts (Fortin et al. 2004, Kubo et al. 2005, which are known to be a signaling platform for cellular functions such as signal transduction, membrane trafficking and cytoskeletal organization. These data indicate that -synuclein plays important roles in vesicle sorting and regulation of catecholamine metabolism in dopaminergic neurons.

LRRK2
LRRK2 mutations were identified as the causative gene for PARK8-linked familial PD (Paisan-Ruiz et al. 2004), (Zimprich et al. 2004). Some screening studies reported that LRRK2 mutations were identified not only in familial PD but also in sporadic PD. Since then, LRRK2 mutations seem to be the most frequent cause of autosomal dominantly-inherited familial PD (Lesage et al. 2006, Ozelius et al. 2006. R1628P and G2385R are polymorphic mutations and have been demonstrated as risk factors for sporadic PD in Asian populations (Di Fonzo et al. 2006, Funayama et al. 2007). The clinical features of patients with LRRK2 mutations essentially resemble sporadic PD with good response to levodopa. Pathological studies of autopsy cases with I1371V, A1441C, Y1699C, G2019S, and I2020T mutations have revealed neuronal cell loss accompanied by Lewy bodies indistinguishable from those of sporadic PD. Some individuals display pleomorphic pathologies, including synucleinopathy, tauopathy, TDP-43 proteinopathy, substantia nigral neuronal loss alone, and neuronal loss with nuclear ubiquitin inclusions (Zimprich et al. 2004, Covy et al. 2009). Although some antibodies against the functional domain fail to detect Lewy bodies, other antibodies directed against the N-terminal and C-terminal residues are able to stain Lewy bodies. Thus, there is controversy as to whether LRRK2 is a component of Lewy bodies. LRRK2 also colocalizes with tau-positive inclusions in fronto-temporal dementia with MAPT N279K mutation (Miklossy et al. 2007). Considering this pleomorphic pathology, LRRK2 seems to be involved in disease-modifying pathways in various neurodegenerative diseases. The LRRK2 protein is a 2527 amino acid polypeptide (~280 kDa), consisting of leucine-rich repeats (LRR), Ras of complex proteins (Roc) followed by the C-terminal of Roc (COR), mitogen-activated protein kinase kinase kinase (MAPKKK) and WD40 domains (Mata et al. 2006). LRRK2 protein belongs to the ROCO protein family. LRRK2 proteins localize in the cytoplasm and membranous organelles, including ER, Golgi apparatus, early endosomes, lysosomes, synaptic vesicles, mitochondria, and the plasma membrane (Hatano et al. 2007). We found LRRK2 bound to lipid rafts within the synaptosomes (Hatano et al. 2007). Several results suggest that LRRK2 also interacts with the Rab family and presynaptic proteins, and modulates the membrane trafficking system.

6
An in vitro phosphorylation assay using thin-layer chromatography revealed that LRRK2 might also function as a serine/threonine kinase (West et al. 2007). The frequent LRRK2 G2019S and adjacent I2020T mutations are located within the kinase domain and exhibit increased activity , Gloeckner et al. 2006. The kinase activity could be regulated by GTP via the intrinsic GTPase Roc domain (Smith et al. 2006, West et al. 2007. Considering that at least some of the mutations alter the kinase activity and are neurotoxic, misregulated kinase activity may explain the core damaging effect of LRRK2 in neurons. Actually, several groups reported that LRKK2 is able to modulate several signal transduction pathways, such as the ERK1/2, TNFα/FasL and Wnt signaling pathways (Berwick & Harvey 2011), and the microRNA pathway, which regulates protein synthesis (Gehrke et al. 2010). The interaction between LRRK2 and parkin is also reported, but there is no evidence to associate LRRK2 with other genes, such as -synuclein, tau, and DJ-1 . Further investigations may reveal how LRRK2 participates in pathways of other PARK gene-related dopaminergic neuronal degeneration.

PINK1/Parkin pathway
Mutations in parkin were identified as the cause of autosomal recessive early onset PD (EOPD) (Kitada et al. 1998), followed by mutations in DJ-1 (Bonifati et al. 2003), and PINK1 (Valente et al. 2004, Hatano et al. 2004). PARK2-linked EOPD is characterized by a spontaneous improvement after sleep or a nap, diurnal fluctuation, some dystonic features which are predominantly in the foot, good response to levodopa, early onset (average age of onset is in the twenties), no dementia, no autonomic failure, and lack of Lewy bodies (Yamamura et al. 1973, Mori et al. 1998. The clinical features of PARK2-linked EOPD are distinguishable from those of patients with the common form of PD. Pathological studies in patients with PARK2-linked EOPD have shown that the substantia nigra and locus coeruleus exhibit selective degeneration with gliosis (Mori et al. 1998). Other pigmented neurons in these regions contain traces of melanin pigments, and the cell cytoplasm is scanty compared with that of normal individuals without neurodegenerative disorders. Although the brain tissue of two cases, which were positive for parkin mutation showed the presence of Lewy bodies (Farrer et al. 2001, Pramstaller et al. 2005, they are generally absent in PARK2-mutation brains. Parkin protein is linked to the ubiquitin-proteasome pathway as a ubiquitin ligase (E3) (Shimura et al. 2000). Most Lewy bodies also stain strongly for ubiquitin, which is highly involved in the protein degradation system. Furthermore, the collaboration between parkin and PINK1 could be associated with mitochondrial quality control systems (Geisler et al. 2010, Vives-Bauza et al. 2010, Narendra et al. 2010, Kawajiri et al. 2010, Matsuda et al. 2010. Parkin-deficient Drosophila exhibit mitochondrial structural alterations in testes and muscle tissue (Greene et al. 2003). PINK1-deficient Drosophila also exhibited degeneration of flight muscles and dopaminergic neuronal cells accompanied by mitochondrial abnormality, and share phenotypic similarity with parkin-deficient Drosophila. Overexpression of parkin reversed the muscle damage in PINK1-deficient Drosophila, but PINK1 overexpression could not reverse the parkin-null-linked damage (Clark et al. 2006, Park et al. 2006. Furthermore, mitochondrial abnormalities in PINK1-or parkin-deficient Drosophila were reversed by knockdown of Mitofusin (Mfn), Optic atrophy 1 (Opa1), or overexpression of Dynamin-related protein1 (Drp1), which participate in mitochondrial fusion and fission (Deng et al. 2008). These findings suggest that PINK1 might associate upstream of parkin to regulate 7 mitochondrial dynamics. The interaction between PINK1 and parkin was also analyzed by using a cultured cell system (Shiba et al. 2009). Narendra and colleagues (2010) reported accumulation of parkin in mitochondria impaired by carbonyl cyanide mchlorophenylhydrazone. Such accumulation promotes the selective autophagy of damaged mitochondria (mitophagy). Several groups have recently shown that PINK1 might be required for parkin-mediated mitophagy (Geisler et al. 2010;Vives-Bauza et al. 2010;Narendra et al. 2010;Kawajiri et al. 2010;Matsuda et al. 2010). We reported that parkin relocates cytoplasm to damaged mitochondria in a membrane potential-dependent manner (Matsuda et al.2010). PINK1 plays a critical role in recruiting parkin to mitochondria with low membrane potential to initiate the autophagic degradation of damaged mitochondria. Pathogenic parkin mutations affect mitochondrial localization. Therefore, loss of the association between parkin and damaged mitochondria could lead to nigral cell death in PARK2-linked EOPD. Immunoprecipitation using anti-LC3 antibodies revealed the direct interaction between PINK1 and endogenous LC3 (Kawajiri et al. 2010). These results suggest that mitochondrial degradation by the PINK1/Parkin pathway is dependent on mitochondrial autophagic activity. Geisler et al. (2010) also reported that the kinase activity of PINK1 in mitochondria might be needed to induce translocation of parkin to depolarized mitochondria. Parkin modified VDAC1 via Lys27 poly-ubiquitylation, and promoted autophagic degradation of damaged mitochondria. Interestingly, they also revealed that the autophagic adaptor p62/SQSTM1 is recruited to mitochondrial clusters and could be associated with PINK1/Parkin-mediated mitophagy. Funayama et al. identified PD patients carrying both parkin and PINK1 mutations, whose onset age was younger than that of patients with the same parkin mutation alone. The presence of digenic mutations, such as parkin plus PINK1 mutations, suggests that a PINK1 mutation could influence the clinical symptoms of parkin-linked PD (Funayama et al. 2008). However, Kitada and colleagues (Kitada et al. 2009) reported that triple knockout of Parkin, PINK1, and DJ-1 resulted in normal morphology and numbers of dopaminergic and noradrenergic neurons in the substantia nigra and locus coeruleus. Therefore, understanding the PINK1/Parkin pathway underlying the mitochondrial dynamics helps clarify the pathomechanics in PD.

Conclusion
The pathological mechanisms and causes of PD remain largely unknown. Most PD patients have not been associated with a particular genetic background or certain exposure of environmental toxins. However, Japanese and European groups performed genome-wide association studies and revealed that the -synuclein locus, the MAPT locus, the LRRK2 locus, the BST1 locus, and 1q32 (PARK16) may be genetic risk factors for PD (Satake et al. 2009, Simon-Sanchez et al. 2009). Although the -synuclein locus, LRRK2 locus, and PARK16 are confirmed as genetic risk factors for PD across populations, BST1 and MAPT may be risk loci correlated with ethnic differences. Conversely, -synuclein is involved in the virulent effects of the environmental toxin MPTP. High mRNA and protein levels ofsynuclein have been described in the brains of MPTP-treated mice (Vila et al. 2000). Although -synuclein-deficient mice exhibited marked resistance to MPTP-induced degeneration of dopamine neurons (Dauer et al. 2002), some lines of -synuclein transgenic mice have exhibited increased sensitivity to MPTP (Nieto et al. 2006, Yu et al. 2008. These results also suggest that various genetic-environmental interactions may have influences on dopaminergic neuronal metabolism and kinetics, such as signal transduction, vesicle www.intechopen.com Etiology and Pathophysiology of Parkinson's Disease 8 transport, autophagy (mitophagy), and mitochondrial stress (Figure 2). In addition, a prionlike transmission of -synuclein may trigger the development of nigral degeneration in PD. Taken together, the new findings based on the interaction between familial PD-related proteins and environmental factors could shed light on the pathomechanisms for PD. Fig. 2. Summary of pathogenetic factors involved in Parkinson's disease (PD). Recent evidence indicated that the PINK1/Parkin pathway participates in mitochondrial removal via autophagy (mitophagy). LRRK2 may associate with cell signaling pathways and the membrane trafficking system. Several environmental factors, such as MPTP, paraquat, and rotenone are known as mitochondrial toxins. In addition, a prion-like transmission ofsynuclein may trigger the development of nigral degeneration in PD.

Acknowledgments
This study was supported in part by grants for research on diseases of the brain from the Ministry of Education, Science and Sports in Japan.