Etiology of Parkinson's disease: Genetics and environment revisited

Idiopathic Parkinson's disease (PD) is the second most common neurodegenerative disorder and affects more than 1 million Americans over the age of 55. The disease selectively affects dopaminergic neurons of the substantia nigra pars compacta, culminating in their demise. After ≈50% of the dopamine neurons and 75–80% of striatal dopamine is lost, patients start to exhibit the classical symptoms of PD including bradykinesia, postural reflex impairment, resting tremor, and rigidity (1, 2). Although there are several treatments that are effective for a number of years, their usefulness wanes over time and is accompanied by unacceptable side effects. New therapies are clearly needed for this disorder.

Despite many years of focused research, the causes of this disease remain to be elucidated. Understanding the cause of PD is critical as that knowledge could lead to directed research that will develop new and potent therapies. The relative contributions of genetic versus environmental factors regarding the cause of PD have been hotly debated. In an attempt to define a cause for this disease, early epidemiological studies examining twins suggested an absence of genetic factors (3). However, these studies were not definitive and could never account for differences in disease progression between twin pairs that could account for discordance in diagnosis. The discovery that the protoxin n-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) causes parkinsonism in both humans and nonhumans further strengthened the hypothesis that PD had an environmental etiology. Other environmental toxins have been shown to induce a parkinsonian state as well, supporting this view (4). However, the recent discovery of inherited forms of PD shifted the emphasis back to genetic factors. Among the different genetic forms of PD, mutations in the gene encoding for α-synuclein have received the most attention. Mutations in this gene cause rare forms of PD. Furthermore, α-synuclein is also present in the Lewy bodies that are the pathoneumonic feature of PD. The outstanding and comprehensive paper by Dauer et al. (5) published in this issue of PNAS examines the nigrostriatal system of α-synuclein knockout mice in response to dopaminergic neurotoxins and re-emphasizes the contributory roles of both genetics and environment in the manifestation of experimental PD. In this study, the authors conclusively demonstrate that targeted disruption of the α-synuclein gene confers specific resistance of dopaminergic neurons to the toxic effects of MPP+ or MPTP in vitro and in vivo. This resistance to degeneration is not mediated through the dopamine transporter but rather through the inability of the toxin to inhibit complex 1 activity in the absence of α-synuclein. Although not the first to do so (e.g., ref. 6), this landmark study provides important evidence that α-synuclein may be critical to dopamine neuronal viability.

Genes Implicated in PD

Although they only occur in a small number of patients, the discovery of genetic forms of PD demonstrated conclusively that PD can occur through inheritance, which has opened a new and exciting area of research. To date, three genes have been found to be associated with inherited PD: α-synuclein, parkin, and UCH-L1 (7–9); four additional loci also have been described (Table 1).

Table 1.

Genes and loci associated with inherited PD

Gene	Locus	Inheritance	Phenotype	Lewy bodies
α-synuclein	4q21-q23	Autosomal dominant	Slightly early onset	+
Parkin	6q25.2-q27	Autosomal recessive	Juvenile onset	−
UCH-L1	4p14-15.1	Autosomal dominant	Typical PD	?
Loci
? (PARK3)	2p13	Autosomal dominant	Typical PD	+
? (PARK4)	4p15	Autosomal dominant	PD/essential tremor	+
? (PARK6)	1p35-p36	Autosomal recessive	Early onset	?
? (PARK7)	1p35-p36	Autosomal recessive	Early onset	?
SCA3	14q32.1	Apparent autosomal dominant	Apparent typical PD	?
SCA2	12q23-q24.1	Apparent autosomal dominant	PD/ataxia	?

Adapted from ref. 10.

α-Synuclein: Structure and Function

α-Synuclein is a member of a small family of proteins that are expressed preferentially within the substantia nigra (11). Its main function has not been fully elucidated. It is known that α-synuclein has a role in synaptic plasticity, as determined from song learning in the zebra finch (12), and in the regulation of dopamine vesicle release, as determined from knockout mice (13). Also, it was hypothesized to function in the transport of fatty acids in the cytoplasm of neurons (10). Despite the lack of knowledge about the functions of this protein, its importance in PD came from the discovery that it is the main component of Lewy bodies. In Lewy bodies, α-synuclein assumes a fibrillary, β-pleated sheet conformation and binds other proteins such as synphilin-1 (14), parkin (15), and the antiapoptotic chaperone 14–3-3 (16). Its conformation can range from unfolded (oligomeric) in solution, to α-helical in the presence of lipid-containing vesicles, to β-pleated sheet or amyloid structure in fibrils. The process of fibrilization is accelerated by nucleation, the A53T mutation, and reactive oxygen species (ROS) (17, 18).

Role of α-Synuclein in Neurodegeneration: Overexpression Models

A critical question that remains to be conclusively answered is whether alterations in one or more of these genes is sufficient to induce degeneration of dopaminergic substantia nigra neurons by itself or whether neural degeneration requires both gene mutations and environmental insults (Table 1). There is evidence for both scenarios. In a study by El-Agnaf et al. (19), α-synuclein expression in cultured neuroblastoma cells induced apoptosis. Because the fibrillary, β-pleated sheet conformation of α-synuclein was observed in the cells, the authors concluded that this form was responsible for neuronal damage. Other in vivo model systems also indicate that α-synuclein overexpression alone can cause nigrostriatal degeneration. An intriguing fly model of PD supports the concept that high levels of α-synuclein can induce neural degeneration. The flies demonstrate a loss of dopamine neurons with advancing age, display Lewy bodies, and most critically, display motor impairments (20).

Transgenic mice have been created to overexpress WT and mutant forms of α-synuclein with variable results. In one study, mice were engineered to overexpress WT α-synuclein by using the platelet-derived growth factor β promoter (21). These animals showed a decreased concentration of dopamine and tyrosine hydroxylase in the striatum and impaired performance in rotorod testing. However, no α-synuclein fibrils and no loss of dopamine neurons were observed. This finding could suggest that environmental triggers against the pathological backdrop of at least WT α-synuclein may be required for actual cell death to occur. Overexpression of A53T mutant α-synuclein in transgenic mice driven by the prion-related protein promoter has generated middle- to late-onset motor deficits, paralysis, and death caused by dehydration and inability to feed. No cell death but neuronal intracytoplasmic fibrillary aggregations were detected in the neuraxis (22, 23).

In contrast, other researchers using the tyrosine hydroxylase promoter to deliver α-synuclein failed to demonstrate consistent neuropathological changes (24–27). In another study, A30P mice showed diffuse aggregates but those aggregates lacked ubiquitin pathology (23).

In addition to transgenic approaches to increase expression of α-synuclein, gene delivery methods have been used. Those delivery methods have the advantage over transgenics in that overexpression of a particular gene is site-specific. Further, the overexpression can occur in adulthood and will not have potential problems of compensation for transgenic manipulations performed during development. In one study, adeno-associated virus (AAV) was used to create transgenic rats that overexpressed A30P α-synuclein. This treatment led to Lewy neurites, fibrillar α-synuclein accumulations, and a 53% loss in dopamine neurons, but no motor impairment (28). In contrast, Björklund and coworkers (29) found only a transient alteration in dopaminergic phenotype and nigrostriatal axonal pathology in rats receiving AAV delivery of A53T α-synuclein. Lo Bianco et al. (6) used a lentiviral system to transfect rats with human WT or mutant A53T and A30P α-synuclein. A selective loss of dopamine neurons was observed with human mutant synuclein. Although the mechanism of cell death has yet to be elucidated, it appears that the fibrils are not responsible because rats transfected with rat α-synuclein display aggregates of α-synuclein fibrils but not neuronal death (6).

Like the Dauer et al. study (5), other investigators have used knockout mice to study the role of α-synuclein. α-Synuclein knockout mice have decreased striatal dopamine function and reduced locomotor response to amphetamine. The animals also showed an increased dopamine release with paired-pulse stimuli. These findings suggest that the mutations in α-synuclein that cause PD are not caused by a loss of function but rather by gain-of-function mutations, and that the role of α-synuclein is a presynaptic regulator of dopamine release (13). The Dauer et al. study extends these original observations to illustrate that α-synuclein is required for neural degeneration to take place after administration of parkinsonian toxins such as MPP+ and MPTP, perhaps through its role in regulation of dopamine release.

Selective Degeneration of Dopaminergic Neurons

Based on the genetics of familial PD and observations in experimental models of overexpression, it is clear that a-synuclein is a modulator of cell death in nigral dopamine neurons. However, several aspect of α-synuclein biology need further examination for a better understanding of the importance of genetics and its interaction with environment. Two key issues that will be of critical importance in furthering our understanding of this matter are: (i) understanding the specific vulnerability of the dopamine-containing neurons, and (ii) determining the role of the various physical forms of α-synuclein in the well-being of dopamine neurons.

Selective Vulnerability

The specific vulnerability of the dopamine-containing neurons in PD is a critical clue for understanding the etiology of this disease and the framework for the vast majority of therapeutic interventions. Researchers have hypothesized that selective vulnerability may result from the (i) preferential expression of parkin and α-synuclein in these neurons, (ii) an absence of calbindin D_28K in dying cells, and/or (iii) dopamine transporter that serves to bring neurotoxins such as MPP+ into cells. During dopamine metabolism, ROS are formed in high concentrations. Of the many pathways that can generate oxidant stress in the breakdown of dopamine, that involving hydrogen peroxide production is particularly damaging. Hydrogen peroxide is generated, then converted to either hydroxyl radical via a Fenton reaction or to superoxide anion and peroxynitrite. The ROS activate mitogen-activated protein kinase and c-Jun N-terminal kinase, which leads to the liberation of cytochrome c from mitochondria and serial activation of caspase 9 and caspase 3, resulting in apoptosis (10).

Dopamine Hypothesis of α-Synuclein Toxicity in PD

Although the biochemistry of toxic dopamine metabolites has been known for many years, their contributory role in human neurodegenerative disease is controversial. With the surge of recent studies investigating the role of α-synuclein in neurodegeneration in PD, new and compelling evidence has implicated a complementary role of α-synuclein, dopamine, and ROS in the selective degenerative mechanism of nigral dopamine neurons in PD.

Xu et al. (30) demonstrated that elevation of α-synuclein in transfected human cultured fetal dopaminergic neuron, but not cortical neurons, results in apoptotic cell death that was prevented by dopamine synthesis inhibition. α-Synuclein was determined to render endogenous levels of dopamine toxic by potentiating the generation of ROS.

Information from several laboratories is emerging that indicates an interaction between dopamine and α-synuclein that causes cell death (6, 30–32). One compelling hypothesis based on data from Conway et al. (31) is that dopamine can result in a kinetic stabilization of the α-synuclein protofibril. Their data show that dopamine-derived quinones can form a dopamine α-synuclein monomer that can inhibit α-synuclein fibrilization. Some researchers believe that oligomeric fibrilization intermediates (protofibrils), rather than the fibrils themselves, are pathogenic. Interestingly, both the A53T and A30P forms of PD accelerate the formation of nonfibrillar, oligomeric protofibrils in vitro. Further, A30P inhibits the conversion of protofibrils to fibrils (17). This finding in familial PD suggests that protofibrils may be pathogenic and fibrils inert, or perhaps even protective (31, 33).

Whether it is the fibrils or some oligomeric intermediates that lead to cell death is still debated. Lashuel et al. (34) have demonstrated that A30P and A53T protofibrils as well as protofibrils implicated in other neurodegenerative diseases (Alzheimer's disease and Huntington's disease) may produce toxicity through the formation of pore-like structures that can permeabilize cell membranes. Circular and elliptical rings corresponding to protofibrils were visualized by atomic force microscopy in a mixture of A53T and WT α-synuclein (35). These annular species resembled pore-forming toxins of the bacteria such as Clostridium perfringens (34). This finding suggests that toxicity may arise from a common structural feature of fibrilization intermediates. The pore- or channel-like activity of protofibrils could be responsible for, or at least contribute to, disease pathology by permeabilizing cell membranes and causing leakage of cellular contents that could lead to cell death.

Interaction Between Genes and Environmental Insults

α-Synuclein monomers are known to aggregate into protofibrillary oligomers and later into fibrils and Lewy bodies. This process of fibrilization is accelerated by nucleation, iron and copper (36), the A53T mutation (18, 37), and the presence of ROS. It is possible that toxins such as MPTP impair mitochondrial function and thus lead to an increase in the concentration of radicals in the neurons, especially if they metabolize dopamine. In the highly oxidative environment found in dopaminergic neurons, the elevated levels of ROS lead to further α-synuclein oligomerization, which causes the demise of the dopamine neurons via the toxic oligomeric species. The oligomers are thought to further lead to generation of more ROS, thus creating a vicious cycle of death for the dopaminergic neurons of the substantia nigra pars compacta (38). This may be one link between the genetic and environmental causes of PD.

Final Comments

This interesting paper by Dauer and coworkers (5) clearly shows that the absence of α-synuclein prevents the damage to dopaminergic nigrostriatal neurons engendered by MPTP or MPP+. They propose that it does so by preventing MPP+ from inhibiting complex 1 activity. Within the framework of MPTP toxicity, inhibition of complex 1 is upstream from a number of critical points that can contribute to cell death, including reductions in energy production and dopamine release. What makes this research even more exciting is the implication that α-synuclein plays a critical role very early in the cascade of degenerative events. These data further the concept that future research should be focused on the role of α-synuclein in the pathology of dopaminergic nigrostriatal neurons and provides hope that a better understanding of these degenerative events will lead to potent methods to prevent the demise of this vulnerable and critical population of neurons.