Using a yeast model to screen for genetic modifiers of α-synuclein toxicity, Susan Lindquist, of MIT, and colleagues have uncovered unexpected links between α-synuclein, the PD gene PARK9, and metal toxicity. The work, published February 1 in Nature Genetics, shows that the yeast homolog of PARK9 can protect the cells from α-synuclein-induced cell death. The same protective function is seen with human PARK9 in mammalian neurons. PARK9 also acts to protect yeast from manganese toxicity, an environmental insult that causes a Parkinson-like disease among miners and metalworkers. The results show for the first time a function for PARK9, and place it and α-synuclein in a highly conserved network of interactions that affect cell survival.

“We’ve managed to connect together three things where previously there was no evidence of a connection,” Lindquist told ARF. “α-synuclein accumulates in the brain of people with PD, and is a defining feature of the disease. We also know that manganese is very toxic to neurons and can cause a syndrome that is kind of like PD, and that mutations in this protein called PARK9 can lead to PD. Those could have been completely independent ways of damaging the neurons, but what we’ve shown is that PARK9 is intimately involved in the biology of the α-synuclein protein and manganese toxicity.”

The Lindquist lab has led the way in using yeast to examine synuclein toxicity as a simple model for the pathogenic processes at play in PD (see ARF related news story). Expression of mutant α-synuclein in yeast causes many of the same problems as in neurons, they have shown, including retardation of protein processing in the endoplasmic reticulum (ER) and Golgi pathways, protein aggregation, cell death. Overexpression of proteins that increase forward transport through the ER/Golgi, most notably the GTPase Ypt1p, can overcome α-synuclein pathology (see ARF related news story).

The human PARK9 gene (also known as ATP13A2) encodes a lysosomal ATPase, and mutations that destroy its function cause a recessive autosomal form of inherited PD (see ARF related news story). In the new study, joint first authors Aaron Gitler, Alessandra Chesi, and Melissa Geddie show that the yeast PARK9 homologue, YPK9, suppresses α-synuclein toxicity without affecting overall levels of α-synuclein protein. YPK9 did, however, affect synuclein localization, normalizing distribution in the plasma membrane and reducing inclusions. This effect resulted from rescue of the vesicle trafficking defect previously described in α-synuclein-expressing yeast.

The effect of PARK9 was conserved in Caenorhabditis elegans and mammalian cells, where the worm PARK9 homologue rescued age-dependent loss of dopamine neurons induced by α-synuclein overexpression. In rat midbrain primary neurons, human PARK9 expression prevented the death of dopaminergic neurons by the A53T mutant of α-synuclein.

Next, the researchers turned back to yeast to probe the function of the protein. The wild-type YPK9 protein localized to vacuole membranes, the yeast equivalent of the lysosome, where PARK9 is found in human cells. Introduction of PD-causing mutations into the yeast protein lowered its expression, altered its distribution, and destroyed its ability to protect against α-synuclein toxicity. The results suggest that human mutations give rise to loss of function as expected from their recessive heritance pattern. YPK9 mutants lacking ATPase activity localized properly to vacuoles, but could not protect cells from α-synuclein toxicity.

By sequence, YPK9 looks like cationic metal transporters, so the researchers tested yeast knockouts for sensitivity to a range of metals and chelators, revealing that the cells were particularly susceptible to manganese toxicity. Overexpression of the wild-type yeast YPK9, but not the disease-associated mutants, protected cells against manganese and the protection required ATPase activity. Because of the high degree of similarity between YPK9 and PARK9, it is likely that the human protein has the same function to protect cells from excess manganese exposure, Lindquist says, but that remains to be proven in human cells.

The yeast results support the idea that multiple pathways that trigger PD may converge on α-synuclein. It is not clear from the current work how manganese sensitivity, PARK9 activity, and α-synuclein toxicity come together in people with PD, but Lindquist hopes that the yeast system will help to clarify that. “It can tell us these three things are related to each other, but it doesn’t tell us exactly how,” she said. “It seems as though working with yeast, you should be able to figure it out.”

Yeast are proving a fertile field for discovering PD-related genes. In addition to YPK9, the current report reveals five other yeast genes, from a variety of pathways, that modify α-synuclein toxicity. Of the five, four had human homologues that protect against α-synuclein toxicity in primary rat midbrain neurons, suggesting they may be important in PD. Lindquist noted that the lab has now screened all 5,000 yeast genes for their effects on α-synuclein toxicity, and will publish the results of the entire analysis shortly.

The approach may also be useful in other neurodegenerative diseases. Lindquist says the lab is now starting to look at Aβ toxicity as a model for Alzheimer disease. “We’re starting now to express the Aβ peptide in yeast cells in such a way that it confers toxicity. We do not yet know if that toxicity is related to the kind of toxicity you see in AD. We’ll have to test in neurons before we know that.”

Signaling Side Effects
Moving to the clinical end of the research spectrum, a study from Ann Graybiel and colleagues, also at MIT, looks at a critical issue in treating the dopaminergic deficits that arise when neurons die in PD. The mainstay of therapy, dopamine replacement with L-dopa, works for a while, but eventually patients develop jerky movements of the arms or head because of treatment. Using a rat model of PD, Graybiel and coworkers have found that these dyskinesias are associated with changes in the expression of striatal-enriched regulatory proteins of the Erk kinase pathway. Because the proteins are enriched in the striatum, they may make better targets for managing dyskinesia than the more widely distributed Ras/Mek/Erk proteins. Their paper appeared in the January 12 issue of PNAS.

In the study, first author Jill Crittenden and coworkers studied rats after unilateral striatal injection of the neurotoxin 6-hydroxydopamine, followed by L-dopa replacement therapy. All of the animals displayed dyskinesias, and the movement problems were accompanied by downregulation of the guanine nucleotide exchange factor CalDAG-GEFI and increased expression of CalDAG-GEFII, as measured by both immunostaining and RT-PCR. The extent of dyskinesia correlated strongly with the changes in gene expression. The two factors were seen to be expressed in different striatal regions and are known to regulate Erk activity in opposite directions. Because of that, the authors suggest that their imbalance may lead to L-dopa-related dyskinesias, which are known to depend on Erk activation.—Pat McCaffrey.

References:
Gitler AD, Chesi A, Geddie ML, Strathearn KE, Hamamichi S, Hill KJ, Caldwell KA, Caldwell GA, Cooper AA, Rochet JC, Lindquist S. alpha-Synuclein is part of a diverse and highly conserved interaction network that includes PARK9 and manganese toxicity. Nat Genet. 2009 Feb 1. Abstract

Crittenden JR, Cantuti-Castelvetri I, Saka E, Keller-McGandy CE, Hernandez LF, Kett LR, Young AB, Standaert DG, Graybiel AM. Dysregulation of CalDAG-GEFI and CalDAG-GEFII predicts the severity of motor side-effects induced by anti-parkinsonian therapy. Proc Natl Acad Sci U S A. 2009 Jan 26. Abstract