Parkin and Its Substrates

See allHide authors and affiliations

Science  13 Jul 2001:
Vol. 293, Issue 5528, pp. 224-225
DOI: 10.1126/science.1063286

Parkinson's disease (PD) has long been considered a textbook example of a sporadic neurodegenerative disorder. Patients with PD have characteristic motor deficits caused by loss of dopaminergic neurons in the brain's nigrostriatal pathway. Postmortem brain tissue from PD patients reveals the presence of inclusions called Lewy bodies in dopaminergic neurons, although whether these inclusions are a cause or a result of the disease is still unclear (see the figure). The discovery several years ago of gene mutations causing rare familial forms of PD provided the first molecular glimpse of a reason for the selective dopaminergic neuronal loss in this disorder. Missense mutations in the gene encoding the α-synuclein protein were found in families with an inherited autosomal-dominant form of PD (1, 2). Various mutations in the PARKIN gene were discovered in families with a rare autosomal-recessive juvenile form of parkinsonism (AR-JP) (3, 4). It is generally believed that the two familial forms of PD are not connected, and so research on α-synuclein and parkin has proceeded separately. This arrangement, however, is set to change with the article by Shimura and colleagues (5) on page 263 of this issue. Knowing that parkin is an E3 ubiquitin ligase and speculating that parkin and α-synuclein might interact, these investigators now provide provocative evidence that parkin regulates the degradation of an unusual form of α-synuclein through the attachment of ubiquitin. The covalent attachment of ubiquitin to a protein by an E3 ubiquitin ligase (ubiquitination) targets that protein for destruction in the cell's garbage dump, the proteasome (6). The fact that α-synuclein is a substrate for parkin is reminiscent of the relationship between β-amyloid precursor protein (substrate) and presenilin (enzyme), mutant forms of which have been implicated in Alzheimer's disease (7).

Parkin ubiquitinates α-synuclein.

(Left) (A) A brain section from a patient with Lewy body dementia, stained with antibody to α-synuclein to show Lewy bodies (arrows) containing α-synuclein. (Right) (A) Lewy bodies are not found in the brains of AR-JP patients. (B) In AR-JP, parkin (an E3 ubiquitin ligase) is defective, leading to accumulation of its substrates: O-glycosylated α-synuclein (αSp22O-glyc) and the putative G protein-coupled receptor Pael-R. Accumulation of both substrates may result in the selective death of nigrostriatal dopaminergic neurons in the brains of AR-JP patients and the motor deficits associated with this rare juvenile form of PD.

A small phosphoprotein found in neurons, α-synuclein is thought to be involved in synaptic vesicle transport (8). Both ubiquitin and α-synuclein are principal components of Lewy bodies, the brain inclusions characteristic of PD and other diseases associated with α-synuclein defects (9). Normal α-synuclein has a tendency to form fibrils that aggregate into sticky clumps. Neurons could get rid of these potentially toxic aggregates by labeling them with ubiquitin and targeting them for degradation. Evidently, this system fails in patients with PD and AR-JP. Intriguingly, protein aggregates associated with other diseases are able to inhibit the ubiquitin-proteasome system (10). Consistent with the notion that a defect in protein ubiquitination and degradation could be a cause of PD, mutations in the E3 ubiquitin ligase parkin and in UCH-L1, a ubiquitin carboxyl-terminal hydrolase, are associated with parkinsonism (11, 12).

In contrast to brains from patients with sporadic PD, the brains of AR-JP patients do not contain Lewy bodies (see the figure) (13). Shimura et al. surmised that parkin might be required to catalyze the ubiquitination of α-synuclein and that the absence or impairment of parkin would lead to the accumulation of non-ubiquitinated α-synuclein in the brains of AR-JP patients. These researchers were able to coimmunoprecipitate α-synuclein and parkin from healthy human brain tissue, suggesting that these two proteins normally interact in neurons. Unexpectedly, the authors found that the α-synuclein species interacting with parkin was O-glycosylated (had carbohydrate moieties attached to some of its hydroxyl groups) and therefore had a larger molecular weight (22 kD) than unglycosylated α-synuclein (16 kD). In fact, O-glycosylation seems to be a prerequisite for α-synuclein ubiquitination because only the 22-kD form is ubiquitinated. Patients with AR-JP lack parkin activity—deletion mutations result in premature chain termination during protein synthesis, and point mutations affect the binding of parkin to its substrates or its ability to ubiquitinate them. The failure of mutant parkin to ubiquitinate glycosylated α-synuclein means that neurons cannot degrade this form, which consequently accumulates in the brain (see the figure). It is tempting to conclude from these findings that accumulation of glycosylated α-synuclein is directly associated with the loss of dopaminergic neurons in AR-JP patients. Indeed, these results could be extrapolated to explain sporadic PD where the accumulation of α-synuclein and parkin in Lewy bodies suggests that there is a defect in the parkin-mediated α-synuclein degradation pathway.

The Shimura et al. results imply that an inability to degrade glycosylated α-synuclein results in AR-JP and possibly sporadic PD. This may well turn out to be the case, but detecting the accumulation of nonubiquitinated glycosylated α-synuclein in AR-JP brains only provides indirect evidence for such a scenario. Furthermore, as all known AR-JP-associated mutations apparently result in the inactivation of parkin, one would predict that other parkin substrates should accumulate in the brains of these patients as well. In a complementary paper in Cell, Imai et al. (14) report that another parkin substrate Pael-R (parkin-associated endothelin receptor-like receptor), indeed accumulates in the brains of AR-JP patients. A putative G protein-coupled receptor, Pael-R belongs to the “difficult-to-fold” class of transmembrane proteins. Misfolded Pael-R is normally efficiently ubiquitinated by parkin and degraded by the proteasome (see the figure). If parkin is defective, misfolded Pael-R is not ubiquitinated or degraded and accumulates in the endoplasmic reticulum (ER) of the neuron, leading to ER stress and cell death. Intriguingly, dopaminergic neurons in the brain produce large amounts of Pael-R, which may account for the selective loss of dopaminergic neurons in AR-JP patients. Besides Pael-R and glycosylated α-synuclein, the synaptic vesicle-associated CDCrel-1 protein (15) and an uncharacterized 30-kD protein (12) are also ubiquitinated by parkin and would be predicted to accumulate in the brains of AR-JP patients.

The next challenge is to determine the relative contributions of these parkin substrates to dopaminergic cell loss in AR-JP and, more importantly, in sporadic PD. Given that α-synuclein accumulates in large amounts inside Lewy bodies, this protein is likely to be a major contributor to dopaminergic cell death. But it is still not clear how buildup of glycosylated α-synuclein could cause the death of nigrostriatal dopaminergic neurons. Pael-R, on the other hand, has not been identified in Lewy bodies, but is known to be neurotoxic when it accumulates in its misfolded state in the ER. It is possible that the PAEL-R gene could be mutated in some cases of sporadic PD. The combined neurotoxic effects of several parkin substrates that accumulate in neurons because they cannot be ubiquitinated or degraded may cause the selective dopaminergic loss in AR-JP and perhaps also in sporadic PD.


View Abstract

Stay Connected to Science

Navigate This Article