29 May 2009. At the 9th International Conference AD/PD, held this past March in Prague, Thomas Gasser of the Hertie Institute in Tuebingen, Germany, presented data that illustrated two emerging concepts in neurodegenerative disease research (see Part 1 of this series). First, the same gene can cause disease in various guises—as an inherited mutation in some familial cases or as a risk variant in some sporadic cases. Second, the overlap between related neurodegenerative diseases is tremendously large and varied. The overlap cuts across the clinical, pathological, and genetics levels, and tau rears its head again and again.

Gasser presented the results of the largest genomewide association study (GWAS) performed to date on Parkinson disease (PD) patient samples. In PD, geneticists in the past decade have assembled a list of some 15 genetic loci. Even so, the large majority of the genetic variance underlying the total burden of PD remains unexplained. The best-understood gene, α-synuclein, gives rise to rare familial PD when mutated, duplicated, or triplicated. However, at the level of pathology, its Lewy body signature is much more widespread than that (Spillantini et al., 1997). Indeed, even beyond PD, α-synuclein pathology shows up in a range of other α-synucleinopathies (see Part 8 in this series). The genetic contribution of α-synuclein alleles to that larger burden of neurodegenerative disease is largely unknown.

The present GWAS represents a collaboration of an international PD Genetics consortium with Andrew Singleton at the National Institutes of Health in Bethesda, Maryland, and additional groups elsewhere, Gasser said. The scientists studied 463,000 quality-controlled SNPs per sample—a huge number as such studies go—in some 1,700 quality-controlled (qc’d) cases and 4,000 qc’d controls from mostly Caucasian U.S. and German samples in a first stage, and then analyzed the 384 top SNPs from this stage in a second sample of 3,500 cases and 4,200 controls. “This gave us good power to detect common risk factors conferring modest effects,” Gasser said.

Which genes survived statistical correction to remain above the Bonferroni line? Only two did—not LRRK2, not parkin, PINK1, or DJ1. α-synuclein and tau stood out in both stages of the GWAS, Gasser reported. For α-synuclein the result was somewhat expected. It further cemented the relationship in certain key genes between pathogenic mutations, duplication, or triplication causing familial and risk alleles promoting sporadic forms of PD. In contrast, tau’s appearance at first glance may seem an astonishing betrayal by a protein implicated in AD and FTD, not primarily PD. But it shouldn’t, noted other researchers. For example, in some cases of PD, tau pathology is readily apparent, said Michael Schlossmacher of the University of Ottawa, Canada. Suspicions about tau playing a role in the pathogenesis of PD arose years ago (e.g., Golbe et al., 2001; Maraganore, 2001). Since then, smaller genetic studies have implicated tau in PD in such a way that a low-expression variant appears to be mildly protective (see tau on PDGene). In the present GWAS, tau’s genetic effect may have been mediated through mRNA expression levels, Gasser said. And the idea is catching on. German researchers recently proposed a spectrum of tauopathy with parkinsonism (Ludolph et al., 2009), and yet other scientists have blamed certain tau variants for the dementia that develops in many patients with advanced PD (Goris et al., 2007). Greetings from the overlap—again.

“In most GWASs of these neurodegenerative diseases, tau comes up. It is not clear what it does, but one thought is that here it could act as a switch to augment α-synuclein pathology,” commented Douglas Galasko of the University of San Diego, California.

Indeed an old finding from familial AD brought up this connection between tau and α-synuclein, when researchers observed that people who had inherited an autosomal-dominant presenilin or APP mutation at autopsy turn out, besides the expected plaques and tangles, to also have had Lewy body pathology in their brains (Lippa et al., 1998). Researchers increasingly agree that APP or presenilin lies upstream of tau in the AD disease pathway, but tau then appears to somehow fire up α-synuclein, as well. What exactly goes on between the two, or whether the pathways are separate, is anyone’s guess at this point, scientists said.

Beyond pointing an accusing finger at tau, the new GWAS presented by Gasser confirms the association of α-synuclein variants and sporadic PD found in previous, smaller studies. The odds ratio for α-synuclein computes to 1.4 for carriers of one risk allele and to 1.9 for carriers of two; together this could account for some 9 percent of the population risk for PD, Gasser said in his talk. The idea is that certain risk alleles increase α-synuclein protein levels. How that might happen, however, remains unclear as the mechanism appears to be more complicated than simple differences in α-synuclein mRNA levels, according to ongoing research in the laboratory of John Hardy at University College, London.

A flip side to this overlap between neurodegenerative diseases (i.e., one clinical entity—different genes) is that one gene can give rise to different clinical pictures. The new GWAS sheds light on this, too. Consider multiple system atrophy (MSA). People with this severe and mysterious disease deteriorate rapidly from parkinsonism or ataxia (or both), and from concomitant failure of their autonomic nervous system. Their brains show α-synuclein pathology mainly in glial cells (Wenning et al., 2008). MSA appears entirely sporadic, as no familial cases have been described as yet. But it has enough in common with PD that Gasser and colleagues decided to check whether any of the 384 top SNPs from the PD GWAS were associated with MSA. This second genetic study, as well, was large for such a rare disease, first testing the SNPs in 413 cases and nearly 4,000 controls and then replicating the top 10 SNPs from that stage in a second, independent cohort. Fifteen different institutions in Europe and the U.S. contributed samples to this study, Gasser noted. α-synuclein emerged as the only MSA risk gene to survive replication and statistical correction. It also stood through further replication in a third sample of 100 pathology-confirmed cases from the Institute of Neurology in London, UK. For MSA, the odds ratio of carrying a homozygous α-synuclein risk allele may be as high as 6.2, Gasser showed (Scholz et al., 2009).

Overall, then, this GWAS drove home the message that α-synuclein alleles that lead to higher expression or protein levels can increase a person’s risk for PD and at least one other α-synucleinopathy, MSA. Tau, too, plays a genetic role in Lewy body diseases. Gasser estimated that taken together, the genes for α-synuclein and tau may account for a fifth of Parkinson disease cases.

Questions after his talk revolved around the notion of whether two pathologies, when they occur together, might fire each other up to accelerate disease. The GWAS does not directly address this because the tau and α-synuclein risk variants were independent of each other. But the question deserves study, Gasser said. To many neurogeneticists, tau’s hand on the tiller of Lewy body diseases may come as no surprise. Tau has been placed at the center of another neurodegenerative disease spectrum before, namely that from AD to FTD (Dermaut et al., 2005). And just this month, British researchers reported having spotted it even further afield, in an aggressive form of multiple sclerosis (Anderson et al., 2009).—Gabrielle Strobel.

This is Part 2 of a nine-part series. See also Part 1, Part 3, Part 4, Part 5, Part 6, Part 7, Part 8, Part 9.

Reference:
Scholz SW, Houlden H, Schulte C, Sharma M, Li A, Berg D, Melchers A, Paudel R, Gibbs JR, Simon-Sanchez J, Paisan-Ruiz C, Bras J, Ding J, Chen H, Traynor BJ, Arepalli S, Zonozi RR, Revesz T, Holton J, Wood N, Lees A, Oertel W, Wüllner U, Goldwurm S, Pellecchia MT, Illig T, Riess O, Fernandez HH, Rodriguez RL, Okun MS, Poewe W, Wenning GK, Hardy JA, Singleton AB, Gasser T. SNCA variants are associated with increased risk for multiple system atrophy. Ann Neurol. 2009 May;65(5):610-4. Abstract