26 July 2011. Sifting through the sequence of all the coding bits of the genome, two independent groups have identified a new mutation that causes Parkinson’s disease. The mutation sits in the gene for vacuolar protein sorting 35 (VPS35), altering its participation in a pathway that sorts and recycles membrane receptors. Some one in 1,000 people with PD carry the defect, estimate the authors of the papers, published July 15 in The American Journal of Human Genetics. Although the mutation’s effects are unknown, it might disrupt sorting of one or more receptors of particular importance in the dopaminergic neurons affected in Parkinson’s, speculated one author, Carles Vilariño-Güell of the University of British Columbia in Vancouver, Canada. Protein sorting problems are starting to look like a trend in neurodegeneration, with similar systems implicated in Alzheimer’s disease and other conditions, agreed investigators from both groups.
Vilariño-Güell was first author on one of the studies, working with senior author Matthew Farrer at the University of British Columbia. Helming the other group were joint first authors Alexander Zimprich of the Medizinische Universität in Vienna, Austria, and Anna Benet-Pagès from Helmholtz Zentrum München in Neuherberg, Germany. Zimprich and Tim Strom, also from Helmholtz Zentrum München, were co-senior authors. Vilariño-Güell announced his team’s results earlier this year at the March International Conference on Alzheimer’s and Parkinson’s Diseases (see ARF related news story).
Both teams used similar exome sequencing strategies to pinpoint the VPS35 mutation, an aspartic acid-to-asparagine, at position 620. Commercial exome selection kits come with nucleic acid probes of known sequences that align with only 1 percent of the genome-encoding proteins, miRNAs, or known regulatory elements such as promoters and 3’ untranslated regions (Bras and Singleton, 2011). These probes are hooked to magnetic beads. When mixed with sheared genomic DNA, the beads hybridize to the matching bits. Then the scientists use magnets to pull out only probe-bound, exomic fractions of the genome. The sequencing is faster and cheaper than tackling an entire genome. Within just a few days, Zimprich said, the researchers have approximately 13 million base pairs to analyze.
Although exome sequencing has only been around for a few years, “it has really taken off as a methodology,” said Andrew Singleton of the National Institute on Aging in Bethesda, Maryland, who was not involved with either study. Scientists have already used the technique to identify rare mutations for a handful of diseases, including amyotrophic lateral sclerosis (see ARF related news story on Johnson et al., 2010).
Like linkage analysis, exome sequencing is best suited to identifying rare but strong mutations in families. But while linkage studies require large families—with perhaps 20 or more affected members—scientists doing exome sequencing can get away with a smaller clan with only four or five ailing relatives, Vilariño-Güell said. Genomewide association studies (GWAS), in contrast, are best suited to discover common, but weak variants associated with idiopathic disease; they require databases with hundreds or thousands of individuals.
GWAS and linkage studies only give a vague address—a linkage group or a nearby single nucleotide polymorphism (SNP)—and they require further steps to pinpoint the actual mutation. One strength of exome sequencing is that, because it reads every base, it leads directly to the causative mutation. However, exome sequencing does have its failings. Some genes do not hybridize well to the selecting probes, Singleton noted, with the result that as much as 20 percent of the exome is only poorly represented in the sequencing. Regulatory sequences outside of the exome, of course, will also be left out. However, exome sequencing is only an intermediate step on the way to whole-genome sequencing gene hunts; some of these problems will disappear once reading the entire genome becomes more affordable.
The Vancouver team started with a Swiss family, while the European group began with an Austrian clan. Based on pedigree analysis, they knew that the disease was inherited in an autosomal-dominant manner, most likely via a single mutation. Carriers of the unknown genetic defect tended to develop Parkinson’s symptoms in their early fifties. Both research groups selected a pair of affected cousins for exome sequencing on the presumption that they inherited the same mutation.
Cousins, of course, share thousands of genetic variants. For that reason, exome sequencing “is not as easy as it sounds,” Vilariño-Güell noted. “There is a lot of follow-up that you need to do to narrow things down.” The researchers took several steps to shorten the lists of potential mutations, discounting variants unlikely to cause the families’ symptoms. For example, some of those thousands of SNPs turned out to be sequencing errors that the Vancouver team was unable to confirm, so they did not pursue them further. Knowing, from pedigrees, that the cousins must be heterozygous for the mutation, the Vancouver group also got rid of any homozygous alleles.
Both teams assumed the mutation they were after was quite rare, so they eliminated from consideration known variants present in databases of common SNPs. They figured a protein-encoding mutation was most likely, so they slimmed down their options by focusing only on variants that would change amino acids or alter splicing. They cut out variants that were not present in other relatives with PD, or that appeared in unaffected family members. Both teams checked additional control cases to identify and remove SNPs carried by anyone who did not have Parkinson’s. In the end, each group narrowed its original lists of thousands down to a single mutation—the VPS35, Asp620Asn. The analysis leading to this mutation is “very convincing,” Singleton said.
VPS35 is part of the retromer, one of the cell’s recycling trucks. Once membrane receptors bind their ligands and are internalized, they wind up in endosomes. Instead of letting those receptors get degraded, the retromer pulls them out and transfers them to the trans-Golgi network, where they are sent right back to the plasma membrane (for review, see Anitei et al., 2010). VPS35 is the central scaffold in the retromer complex, linking to several other members as well as the receptor cargo.
What is the functional consequence of this genetic variation? The Asp620Asn mutation is unlikely to drastically alter the protein’s structure, Vilariño-Güell said. In co-immunoprecipitation experiments, he determined that the mutant VPS35 could still interact with other retromer proteins. He theorizes that the mutation prevents or limits retromer binding to one or more receptors. If those receptors were crucial to dopaminergic neurons, altering their recycling could eventually cause Parkinson’s, he suggested. “I think that will turn out to be true,” agreed Scott Small of Columbia University in New York. Small has studied VPS35 in the Alzheimer’s brain, but was not involved in either of these PD studies.
VPS35 is not the first protein-sorting gene implicated in neurodegeneration, nor is it linked only to Parkinson’s. VPS35 controls amyloid-β levels, and VPS35 protein levels are low in some parts of the Alzheimer’s brain (Small et al., 2005). Similarly, retromer deficiency accelerates pathology in mouse and fly models of Alzheimer’s (see ARF related news story on Muhammad et al., 2008). The sorting proteins SorCS1 and SorL1 are also involved in the processing of amyloid precursor protein (see ARF related news story on Lane et al., 2010).
Another sorting protein, sortilin, has been implicated in frontotemporal lobar dementia (FTLD) because it binds progranulin; mutations in progranulin can cause FTLD (see ARF related news story on Carasquillo et al., 2010 and Hu et al., 2010). Further, mutations in Rab7—which recruits the retromer to the endosome—cause the peripheral neuropathy Charcot-Marie-Tooth disease (Seaman et al., 2009). Mutations in the motor protein dynactin, also required for endosome-to-Golgi transport, cause the rare neurodegenerative condition Perry syndrome (see ARF related news story on Farrer et al., 2009); another dynactin mutation predisposes people to ALS and FTLD (Münch et al., 2005). “Everything seems to be fitting around these retromers,” Vilariño-Güell said.
To Small’s knowledge, all the diseases linked to the retromer are neurological. “When blocked, that particular route, from the endosome back to the Golgi is more likely to affect neurons than other cell types,” he speculated. Theoretically, minor alterations that affect VPS35’s binding to different receptors could explain the range of symptoms that go along with retromer diseases; some receptors might be important in the neurons implicated in Parkinson’s, while others could be key to the nerves that degenerate in Alzheimer’s, Vilariño-Güell suggested.
The discovery leads to two obvious questions. For one, what is the VPS35 mutation actually doing to cause disease? Farrer’s group has already initiated biochemical experiments to ferret out the answer. Second, how common is this mutation? Both groups estimate that approximately one in 1,000 Europeans with Parkinson’s harbor the VPS35 Asp620Asn mutation, but the proportions could be different in other populations. Strom, in conjunction with the International Parkinson Disease Genomics Consortium, is currently screening thousands of samples to come up with a better estimate of the mutation's prevalence.—Amber Dance.
Vilariño-Güell C, Wider C, Ross OA, Dachsel JC, Kachergus JM, Lincoln SJ, Soto-Ortolaza AI, Cobb SA, Wilhoite GJ, Bacon JA, Behrouz B, Melrose HL, Hentati E, Andreas Puschmann A, Evans DM, Conibear E, Wasserman WW, Aasly JO, Burkhard PR, Djaldetti R, Ghika J, Hentati F, Krygowska-Wajs A, Lynch T, Melamen E, Rajput A, Rajput AH, Solida A, Wu RM, Uitti RJ, Wszolek ZK, Vingerhoets F, Farrer MJ. VPS35 mutations in Parkinson’s disease. Am J Hum Genet. 2011 Jul 15;89(1):162-7. Abstract
Zimprich A, Benet-Pagès A, Struhal W, Graf E, Eck SH, Offman MN, Haubenberger D, Spielberger S, Schulte EC, Lichtner P, Rossle SC, Klopp N, Wolf E, Seppi K, Pirker W, Presslauer S, Mollenhauer B, Katzenschlager R, Foki T, Hotzy C, Reinthaler E, Harutyunyan A, Kralovics R, Peters A, Zimprich F, Brücke T, Poewe W, Auff E, Trenkwalder C, Rost B, Ransmayr G, Winkelmann J, Meitinger T, Strom TM. A mutation in VPS35, encoding a subunit of the retromer complex, causes late-onset Parkinson disease. Am J Hum Genet. 2011 Jul 15;89(1):168-75. Abstract