News & Commentary

Putting a PARKINg brake on neurodegeneration

Mutations in the parkin gene are associated with the juvenile-onset form of Parkinson's disease (PD). Recent studies suggest that Parkin plays a central role in a cellular pathway that provides neuroprotective function. Manipulating Parkin activity may prevent neurodegeneration in juvenile onset as well as more common sporadic forms of PD.

Not long ago, neurodegenerative diseases were considered the most obscure and intractable disorders affecting the human brain. The situation has changed rapidly, thanks in a large part to breakthroughs in human genetics that help pinpoint genes associated with familial forms of those diseases such as Alzheimer's disease and PD. PD is the most common movement disorder and the second most common neurodegenerative diseases. PD patients suffer from rigidity, resting tremor, and slowness of movement. The movement disorder is largely attributed to the deficiency of brain dopamine contents resulting from the loss of dopaminergic neurons in the midbrain.

Although most common forms of PD are sporadic, it has long been recognized that genetics plays a role in certain cases. Molecular cloning of genes linked to familial forms of PD has provided initial insights into its pathogenic mechanisms. Missense mutations in α-Synuclein (α-Syn) were first shown to cause rare forms of autosomal dominant PD.1 α-Syn is an abundant brain protein enriched at presynaptic terminals. Mice with α-Syn gene deleted show increased striatal dopamine release, but do not develop PD-like pathological phenotypes, suggesting that α-Syn mutations cause PD through a dominant gain-of-function mechanism. Significantly, wild-type α-Syn protein was found to be a major component of Lewy bodies, the proteinaceous aggregates present in PD and related diseases. This suggests that the accumulation and aggregation of α-Syn is intimately involved in disease pathogenesis. Further support came from transgenic animal studies in which overexpression of wild-type or mutant forms of α-Syn in mouse and Drosophila led to α-Syn aggregation and neuronal dysfunction.2

Mutations in the parkin gene were linked to autosomal recessive juvenile parkinsonism (AR-JP).3 AR-JP patients develop the typical parkinsonian symptoms with attendant loss of midbrain dopaminergic neurons. But this usually occurs without Lewy body pathology. Biochemical studies have demonstrated that Parkin has E3 ubiquitin-protein ligase activity, and that AR-JP-linked parkin mutations abolished this activity. Ubiquitin-protein ligases are components of the ubiquitin–proteasome pathway that degrades substrate proteins with abnormal conformations. In the light of Parkin's biochemical function, a prevailing working hypothesis for the pathogenesis of AR-JP proposes that loss of Parkin function leads to the accumulation of abnormal Parkin substrates, which cause neurotoxicity and dopaminergic neuron death.

A number of Parkin substrate proteins have been identified through in vitro protein–protein interaction studies. These include CDCrel-1, a synaptic vesicle-associated protein; Synphilin-1, a α-Syn interacting protein also localized to the synapse; αSp22, an O-glycosylated form of α-Syn; Pael-R, a GPCR-like seven-transmembrane protein; and cyclin E, which was previously implicated in the regulation of neuronal apoptosis. Studies using cell culture models have provided clues about the potential functions of these substrates in the disease process. For example, coexpression of α-Syn and Synphilin-1 results in Lowy body-like inclusion formation,4 overexpression of Pael-R causes unfolded protein stress in the ER and subsequent cell death,5 and kainate-induced cell death, which is thought to be mediated by the accumulation of cyclin E, can be prevented by Parkin overexpression.6 However, whether these factors play a causative role in the pathogenesis of AR-JP is not known. To address this question, organismal models are needed.

Drosophila has recently been established as a model system for studying human neurodegenerative diseases and mental disorders. For example, genetic studies in Drosophila have provided invaluable insights into the mechanisms of polyglutamine diseases and fragile X syndrome. The recent development of α-Syn transgenic models has validated Drosophila as an excellent system for studying α-Syn-mediated neurodegeneration in PD.7 To study the role of Parkin and its substrates in the pathogenesis of AR-JP, Yang et al used a transgenic approach to demonstrate that pan-neural overexpression of human Pael-R causes selective degeneration of dopaminergic neurons in Drosophila. Overexpression of human Parkin leads to the degradation of Pael-R and suppresses Pael-R neurotoxicity. Conversely, inhibition of endogenous fly Parkin activity exacerbates Pael-R toxicity.8 These results not only established a causative link between Pael-R accumulation and selective dopaminergic neuron loss, but also demonstrated that the fly model could be used to identify endogenous factors that can modify the disease phenotypes. Significantly, the authors showed that overexpression of Parkin can also suppress α-Syn neurotoxicity in Drosophila. Overexpression of Parkin suppresses α-Syn toxicity by mitigating α-Syn-induced neuritic pathology and reducing the aggregated form of α-Syn, and this occurs without significantly changing the overall α-Syn protein level. The authors hypothesize that the toxic form of α-Syn, which is the presumed Parkin substrate, may constitute only a minor portion of the total α-Syn protein pool. Further studies are needed to elucidate the nature of neurotoxic α-Syn species in order to understand the exact molecular mechanism by which Parkin suppresses α-Syn toxicity.

One of the major unresolved mysteries in PD as well as other neurodegenerative diseases is that a selective group of neurons die, despite the widespread expression of the disease-causing genes. This cell-type specificity is recapitulated in both the α-Syn and Pael-R transgenic flies, where ubiquitous expression of these toxic proteins results in selective degeneration of dopaminergic neurons. It is plausible that dopaminergic neurons may have a smaller capacity for handling misfolded proteins or exhibit a higher sensitivity to these toxic proteins, a notion supported by recent cell culture studies.9 It is also possible that dopaminergic neuron-specific factors are involved. Dopamine metabolites are known to cause neurotoxicity through oxidative stress, and environmental toxins that induce oxidative stress by disrupting mitochondrial respiratory chain have been shown to cause parkinsonian symptoms in humans and experimental animals.10 Thus, oxidative stress intrinsic to dopaminergic neurons may act together with other cellular insults to contribute to selective cell death. In this regard, it is interesting to note that the more recently identified PD-causing DJ-1 gene may participate in oxidative stress response.11 The recent demonstration of mitochondrial pathology in the muscle tissue of parkin mutant flies warrant further investigation into the roles of mitochondrial dysfunction and oxidative stress in Parkin-mediated neurodegeneration.12

Genetic studies in Drosophila have established the relationships between known gene products in the disease process. Further studies employing the power of fly genetics promise to delineate the genetic pathway of PD pathogenesis, of which Parkin appears to be a central component. Considering that the accumulation and aggregation of misfolded proteins is a feature shared by many neurodegenerative diseases, understanding Parkin-mediated neurodegeneration may shed light on the pathogenic mechanisms of related diseases. The ultimate goal of such studies is to develop novel and rational strategies to prevent the onset and halt the progression of these devastating diseases. Available data already suggest that manipulating the activity or expression level of Parkin with small molecules may provide a novel mechanism-based therapeutic approach.


  1. 1

    Polymeropoulos MH et al. Science 1997; 276: 2045–2047.

  2. 2

    Dawson T, Mandir A, Lee M . Neuron 2002; 35: 219–222.

  3. 3

    Kitada T et al. Nature 1998; 392: 605–608.

  4. 4

    Chung KK et al. Nat Med 2001; 7: 1144–1150.

  5. 5

    Imai Y et al. Cell 2001; 105: 891–902.

  6. 6

    Staropoli JF et al. Neuron 2003; 37: 735–749.

  7. 7

    Feany MB, Bender WW . Nature 2000; 404: 394–398.

  8. 8

    Yang Y, Nishimura I, Imai Y, Takahashi R, Lu B . Neuron 2003; 37: 911–924.

  9. 9

    Petrucelli L et al. Neuron 2002; 36: 1007–1019.

  10. 10

    Betarbet R, Sherer TB, Greenamyre JT . BioEssays 2002; 24: 308–318.

  11. 11

    Bonifati V et al. Science 2003; 299: 256–259.

  12. 12

    Greene JC et al. Proc Natl Acad Sci USA 2003; 100: 4078–4083.

Download references

Author information

Correspondence to Bingwei Lu.

Rights and permissions

Reprints and Permissions

About this article

Further reading