Like Alzheimer’s, like Parkinson’s? The last decade has seen study after study confirm that falling amyloid-β (Aβ) in the cerebrospinal fluid (CSF) is an indicator of incipient Alzheimer’s disease. One theory is that CSF levels drop as Aβ is sequestered in amyloid plaques in the brain. Might something similar be true for α-synuclein, the major component of the intracellular protein aggregates (Lewy bodies) found in Parkinson’s and related neurodegenerative diseases? In last week’s Lancet Neurology, researchers led by Brit Mollenhauer at the Paracelsus-Elena-Klinik, in Kassel, Germany, report one of the largest studies yet to test this idea. The researchers found that, on average, people with synucleinopathies, including PD, dementia with Lewy bodies, and multiple system atrophy, have lower levels of CSF α-synuclein than normal controls or people with AD. “It is pretty exciting because it starts to put α-synuclein on the map as a biomarker of a distinct pathological process,” Les Shaw, University of Pennsylvania, Philadelphia, told ARF. “I think it is going to generate a lot of interest.” Shaw was not involved in this study.

Numerous studies have looked at CSF α-synuclein in the context of PD, but the results are contradictory. Some found CSF synuclein can distinguish people with PD from controls (e.g., Mollenhauer et al., 2008), while others did not (Reesink et al., 2010). As Lucilla Parnetti, University of Perugia, Italy, lays out in an accompanying Lancet Neurology commentary, reasons for the inconsistencies include small sample sizes and questionable accuracy of methods used. “The large and accurate work by Mollenhauer and colleagues confirms the reliability of their proposed method,” she writes.

Mollenhauer collaborated with Michael Schlossmacher, formerly at Brigham and Women’s Hospital, Boston, and now at the University of Ottawa, Canada, to perfect an enzyme-linked immunosorbent assay (ELISA) for α-synuclein. In initial tests on a cohort of 273 people with diagnoses of Parkinson’s, dementia with Lewy bodies (DLB), multiple systems atrophy (MSA), AD, or other neurodegenerative diseases, this ELISA distinguished the synucleinopathies (PD, MSA, and DLB) based on their consistently lower CSF values. Low CSF synuclein also discriminated DLB from AD in 41 patients with diagnoses confirmed postmortem. In a separate validation cohort of 407 patients, those with PD, DLB, and MSA also had, on average, lower CSF α-synuclein than those with other movement disorders such as progressive supranuclear palsy, or other neurological conditions such normal pressure hydrocephalus. Mollenhauer and colleagues calculated a cutoff value of 1.6pg/microLiter or less as an indicator of PD, DLB, or MSA.

Because of overlap in the data (some individuals with CSF synuclein below the cutoff value did not have a synucleinopathy), the researchers caution that measuring CSF synuclein alone will not enable neurologists to establish a diagnosis. “It is not a slam dunk in discriminating Parkinson’s versus controls, but the trend [for CSF synuclein] is lower, and the utility will be in helping confirm diagnosis,” said Mark Frasier, from the Michael J. Fox Foundation for Parkinson’s research, which supported the study and is spearheading efforts to standardize methodologies for measuring fluid biomarkers. “The major diagnostic utility was the positive predictive value,” said Shaw. “If you do test positive, then you are very likely to have the underlying pathology,” he said.

Which pathology is another question. The study is the first to measure CSF synuclein in multiple systems atrophy, which is often misdiagnosed. It turns out that patients with PD and MSA exhibit very similar CSF profiles, although their clinical symptoms differ. “On one hand, that is disappointing, because it would be nice to have a marker that differentiates between the two,” said Frasier. “On the other hand, this supports the evidence that there is a mechanistic reason for lower synuclein in CSF, since in both disorders there is an accumulation of synuclein in the brain.” α-synuclein deposits occur in different cell types and brain areas in PD and MSA, with the latter showing deposits primarily in oligodendrocytes, not neurons, and in the brain stem and basal ganglia. The data for DLB also support this hypothesis. CSF α-synuclein levels in DLB patients fell between those for PD/MSA and normal controls or people with Alzheimer’s. This fits with DLB not being a pure synucleinopathy, but having neuropathological overlap with AD.

Mollenhauer and colleagues write that including multiple markers could help improve diagnostic potential. In fact, they also looked at CSF tau and Aβ and found that DLB patients had levels of both that fell in between those recorded for AD and PD/MSA patients. In her commentary, Parnetti notes that she and her collaborators, including Omar El-Agnaf at United Arab Emirates University in Al Ain, found that the α-synuclein-to-tau ratio was best able to distinguish Parkinson’s from other neurodegenerative disorders, including DLB and AD (their paper is in press). Whether other markers could be found to distinguish PD from MSA is unclear at present, but Mollenhauer and colleagues suggest that “the joint measurement of oligomeric α-synuclein and as yet undiscovered post-translationally modified α-synuclein species in CSF might allow a more definitive laboratory diagnosis of an α-synuclein-associated disorder in the future.” El-Agnaf, who has had some success developing assays for oligomeric α-synuclein (see ARF related news story) agrees. “We definitely need more than one marker to improve the sensitivity and specificity. That is the key, and that is what we should focus on,” he told ARF.

In that respect, the PD field lags some years behind AD, which has harvested a bounty of biological, neuropsychological, and brain imaging data from the Alzheimer’s Disease Neuroimaging Initiative and other longitudinal aging studies. Going forward, the Parkinson’s Progression Markers Initiative (see ARF related news story), for which Mollenhauer leads a site in Germany, will test longitudinally for changes in a variety of fluid and imaging markers. The PPMI aims to develop standardized test protocols. For example, in another recent study that found lower α-synuclein in the CSF of PD patients, researchers led by Jing Zhang at the University of Washington, Seattle, found that contamination of CSF samples with blood can lead to spurious results (see Hong et al., 2010). This confirms a technical issue Mollenhauer has been pointing out for some years. To get an accurate reading, it is important that the lab technician centrifuge the CSF to remove any blood cells or other cellular contaminants that might release their content into the CSF upon freezing. “Centrifugation may not be critical for Aβ or tau, but it must be done for α-synuclein,” Mollenhauer said.—Tom Fagan


  1. Mollenhauer’s study benefits from examining three cohorts, including an autopsy-confirmed cohort with dementia with Lewy bodies (DLB) patients—a group often difficult to distinguish from Alzheimer’s disease (AD) by clinical testing. Measurements of several markers of neurodegenerative disease were made, including α-synuclein, tau, and Aβ1-42. By combining the cerebrospinal fluid measurements for α-synuclein and tau with age, the synucleinopathy group (Parkinson’s disease, DLB, and multiple systems atrophy) could be discriminated from AD and other neurological conditions more clearly than by α-synuclein alone. The α-synuclein differences were confirmed in the autopsy-confirmed cohort of DLB patients.
    While able to discriminate between conditions whose primary pathology is α-synuclein and those with other types of pathology, discrimination between different types of synucleinopathy has not been achieved. In many ways, this is as important as being able to differentiate between conditions with differing pathological causes, as the initial stages of the diseases may appear to be very similar, and more accurate diagnoses at an early stage could allow for more targeted treatments.

    Discrimination between these conditions could be achieved by a closer analysis of different forms of α-synuclein, focusing on oligomeric α-synuclein, truncated forms, and post-translational modifications, such as phosphorylation and ubiquitination.
    An important fact that can be deduced from previous CSF α-synuclein studies is that variation between cohorts, sample preparation, and antibody choice can have an enormous impact on the results. To gain the full clinical benefits from these findings, the results need to be replicated in larger cohorts, including individuals with early diagnoses of mild synucleinopathies and by different laboratories. Further prospective studies which track the changes in α-synuclein throughout disease progression will also be required. This current study provides a basis for future combination studies of CSF biomarkers in synucleinopathies, and it will be exciting to follow how this field progresses in the coming months

    Emma L. Jones is funded by the Alzheimer’s Society (Grant Number 112) with support from The Henry Smith Charity.

    View all comments by Emma Jones

Make a Comment

To make a comment you must login or register.


News Citations

  1. An Oligomer Test for PD—Could One for AD Be Close Behind?
  2. PPMI: Brain Imaging To Reveal Preclinical Parkinson’s Signature?

Paper Citations

  1. . Direct quantification of CSF alpha-synuclein by ELISA and first cross-sectional study in patients with neurodegeneration. Exp Neurol. 2008 Oct;213(2):315-25. PubMed.
  2. . CSF α-synuclein does not discriminate dementia with Lewy bodies from Alzheimer's disease. J Alzheimers Dis. 2010;22(1):87-95. PubMed.
  3. . DJ-1 and alpha-synuclein in human cerebrospinal fluid as biomarkers of Parkinson's disease. Brain. 2010 Mar;133(Pt 3):713-26. PubMed.

External Citations

  1. Alzheimer’s Disease Neuroimaging Initiative
  2. Parkinson’s Progression Markers Initiative

Further Reading

Primary Papers

  1. . α-Synuclein and tau concentrations in cerebrospinal fluid of patients presenting with parkinsonism: a cohort study. Lancet Neurol. 2011 Mar;10(3):230-40. PubMed.
  2. . Biochemical diagnosis of neurodegenerative diseases gets closer. Lancet Neurol. 2011 Mar;10(3):203-5. PubMed.