
A blood marker for Parkinson’s Disease: Neuronal
exosome-derived α-synuclein
Annika Kluge ( Annika.Kluge@uksh.de )
Department of Neurology, University Hospital Kiel
Josina Bunk
Institute of Biochemistry, Christian-Albrecht-University Kiel
Eva Schaeffer
University Hospital Kiel, Department of Neurology
Alice Drobny
Department of Molecular Neurology, University Hospital Erlangen, Friedrich-Alexander University
Erlangen-Nürnberg
Wei Xiang
Friedrich-Alexander-University Erlangen-Nürnberg (FAU)
Henrike Knacke
Department of Neurology, University Hospital Kiel
Simon Bub
Department of Molecular Neurology, University Hospital Erlangen, Friedrich-Alexander University
Erlangen-Nürnberg
Wiebke Lückstädt
Institute of Anatomy, Christian-Albrecht-University Kiel https://orcid.org/0000-0002-1698-6770
Philipp Arnold
Institute of Functional and Clinical Anatomy, Friedrich-Alexander-University Erlangen-Nürnberg
Ralph Lucius
Institute of Anatomy, Christian-Albrecht-University Kiel
Daniela Berg
University Hospital Schleswig-Holstein Kiel https://orcid.org/0000-0001-5796-5442
Friederike Zunke
University Hospital Erlangen
Article
Keywords: Parkinson's Disease, biomarker, α-synuclein
Posted Date: August 12th, 2021

1
A blood marker for Parkinson’s Disease: Neuronal exosome-derived α-synuclein 1
2
Annika Kluge
1*
, Josina Bunk
2
, Eva Schaeffer
1
, Alice Drobny
3
, Wei Xiang
3
, Henrike Knacke
1
, 3
Simon Bub
3
, Wiebke Lückstädt
4
, Philipp Arnold
5
, Ralph Lucius
4
, Daniela Berg
1#
, Friederike 4
Zunke
3#*
5
6
1
Department of Neurology, University Hospital Kiel, Kiel, Germany 7
2
Institute of Biochemistry, Christian-Albrecht-University Kiel, Kiel, Germany 8
3
Department of Molecular Neurology, University Hospital Erlangen, Friedrich-Alexander 9
University Erlangen-Nürnberg, Erlangen, Germany 10
4
Institute of Anatomy, Christian-Albrecht-University Kiel, Kiel, Germany 11
5
Institute of Functional and Clinical Anatomy, Friedrich-Alexander-University Erlangen-12
Nürnberg, Erlangen, Germany 13
#
contributed equally 14
*correspondence: Annika.Kluge@uksh.de; Friederike.Zunke@fau.de 15
16
Abstract 17
To date, no reliable clinically applicable biomarker has been established for Parkinson’s 18
disease (PD). Our results indicate that a long hoped blood test for Parkinson’s disease may 19
be realized. We here assess the potential of pathological α-synuclein originating from 20
neuron-derived exosomes from blood plasma as a possible biomarker. Following the 21
isolation of neuron-derived exosomes from plasma of PD patients and non-PD individuals 22
immunoblot analyses were performed to detect exosomal α-synuclein. Under native 23
conditions significantly increased signals of disease-associated α-synuclein forms in neuron-24
derived exosomes were measured in all individuals with PD and clearly distinguished PD 25
samples from controls. By performing a protein misfolding cyclic amplification assay these 26
aggregates could be amplified and seeding could be demonstrated. Moreover, the 27
aggregates exhibited β-sheet-rich structures and showed a fibrillary appearance. Our study 28
demonstrates that the detection of pathological α-synuclein conformers from neuron-derived 29
exosomes from plasma samples has the potential of a promising blood-biomarker of PD. 30
31
Keywords: α-synuclein, Parkinson´s disease, biomarker, plasma, extracellular vesicles, 32
neuron-derived exosomes, protein misfolding cyclic amplification. 33
34
Introduction 35
To date the gold-standard to confirm Parkinson’s Disease (PD) is the post-mortem detection 36
of misfolded α-synuclein (α-syn) as structural component of Lewy bodies in dopaminergic 37
neurons in the substantia nigra (SN)
1
. However, in the clinical routine the diagnosis of PD is 38
still based on the detection of motor symptoms, supported by imaging techniques and the 39
assessment of concurrent non-motor symptoms and risk factors for PD
2
. Therefore, the 40
correct diagnosis and appropriate therapy is still highly dependent on the professional 41
experience of the examiner, and many epidemiological or post-mortem studies found high 42

2
rates of misdiagnoses in PD
3-6
. Another major shortcoming of the clinical approach to 43
diagnose PD is the substantially delayed diagnosis in the course of the disease, as the 44
diagnosis-defining motor symptoms occur only late in the neurodegenerative process, i.e. 45
when more than 50 % of dopaminergic neurons in the SN have already been lost
7
. An earlier 46
detection of the disease, ideally in the prodromal phase before motor symptoms occur, is of 47
utmost importance for the development and application of disease-modifying therapies. 48
Finally, the assessment of clinical symptoms by scales is still used as primary outcome 49
parameter in most clinical trials. This semi-quantitative approach is an imprecise reflection of 50
actual disease progression, depending on a variety of potential confounders such as 51
medication intake, examiner’s experience and physical as well as psychological form of the 52
patient on the day of examination. 53
Taken together, there is an urgent need for an objective and reliable biomarker, to improve 54
the diagnostic accuracy of PD, detect the disease in early stages (preferably in the prodromal 55
state) and monitor disease progression. In this respect, the detection of pathological α-syn as 56
neuropathological hallmark of PD has been in the centre of attention in a wide range of 57
studies
8
. Many studies have focused on the identification of α-syn in accessible peripheral 58
tissues for instance biopsies of the gastrointestinal tract, skin or salivary glands
9-12
. 59
Moreover, there are first promising findings regarding the identification and characterization 60
of pathological α-syn forms in biofluids, such as the cerebrospinal fluid (CSF)
13
. However, 61
apart from still highly varying outcomes regarding sensitivity and specificity, all these 62
techniques are limited due to their invasiveness. Compared to those options an easy and 63
low-risk obtainable medium is blood plasma or serum
14
. With regard to contaminations and 64
inconsistent α-syn levels in the blood
14-18
, recent studies focused on extracellular vesicles 65
(EVs) like exosomes (30-100 nm membrane vesicles of endocytic origin) and microvesicles 66
(100 nm-1 µm)
19-21
. EVs released by cells of the central nervous system (CNS) (neuron-67
derived exosomes, NEs) have the capacity to pass the blood brain barrier (BBB) and 68
transport nucleic acids and proteins including α-syn
20,22
. In this manuscript the terms 69
‘exosomes’ and ‘extracellular vesicles’ will be used interchangeably to denote vesicles, which 70
are released extracellularly and can be isolate from plasma samples. For a more detailed 71
description of exosome nomenclature please see the article of the International Society for 72
Extracellular Vesicles
23
. 73
74
In this study we combined a specific preparation of NEs and the use of structure-specific 75
antibodies to detect pathological α-syn conformers isolated from the blood of PD patients and 76
compared them to non-PD individuals. Using a protein misfolding cyclic amplification assay 77
(PMCA)
13
, we analyzed the detected protein form in its capacity to seed α-syn aggregation 78
and built filamentous structures. In this study we demonstrate for the first time that the 79
detection of pathological α-syn conformers extracted from NEs is possible and can be 80
applied as suitable, easy to assess biomarker for PD that reliably discriminates patients from 81
controls. 82
83
Results 84
Demographics 85
Plasma samples were collected from 15 patients with PD (mean age 67 years, range 86
46-84 years) and 15 controls (mean age 75 years, range 50-85 years). There was no age 87
distribution difference among the groups (p = 0.49). Mean disease duration of PD patients 88

3
was 3 (1-13) years and mean clinical motor symptom score (Movement Disorder Society 89
Unified Parkinson's Disease Rating Scale Part 3, MDS-UPDRS-III) was 26.7. Summarized 90
data of both groups are listed in Table 1; available clinical data for each patient/control is 91
listed in the Extended Data Table 1. 92
93
Isolation and detection of EVs from peripheral blood 94
After gradual centrifugation and exosome precipitation (Fig. 1a), the successful isolation of 95
EVs was confirmed through immunoblotting, dynamic light scattering (DLS) and transmission 96
electron microscopy (TEM). The isolation of EVs was confirmed through dot blot analyses 97
using native plasma samples and comparing them to isolated EVs (PD#1-#2, Ctrl#1-#2) 98
(Fig. 1b, Extended Data Fig. 1a). The exosomal marker CD63 was significantly enriched in 99
samples of EVs after normalization to total protein (Fig. 1c). Increased CD63 levels within the 100
samples of EVs indicated a sufficient protocol for enrichment of EVs. Following the exosomal 101
isolation western blot analyses depicted an increased anti-CD63 antibody signal in both 102
groups (PD#1-#5, Ctrl#1-#5) (Fig. 1d). In comparison with untreated plasma samples, 103
increased signal intensities were detected after isolation of EVs (Extended Data Fig. 1b). 104
Comparing control and PD samples, no significant differences in CD63 levels could be found 105
after normalization to Coomassie Brilliant Blue (CBB) staining (Fig. 1e). Further 106
characterization of the diameter and morphology of EVs was gained by negative stain TEM 107
and showed a homogenous preparation of EVs (Fig. 1f). All images can be found in 108
Extended Data Fig. 1c. Particle size measurement from TEM images showed no differences 109
in EVs between PD and control samples (Extended Data Fig. 1d). Furthermore, DLS 110
measurements confirmed the presence of uniform particles with the size as EVs (Fig. 1g)
19
. 111
Analyzed samples of PD patients and controls showed no differences in mean radius 112
distribution according to the TEM-based size distribution. In summary, we demonstrate the 113
sufficient isolation of EVs from blood plasma samples, exhibiting no differences in size or 114
morphology between PD patients and controls. 115
116
Identification of NEs from peripheral blood 117
The purification of NEs (Fig. 2a) led to a significantly increased signal of L1 cell adhesion 118
molecule (NCAM-L1) when compared to native plasma samples and plasma-derived 119
exosomes (PD#4-#5, Ctrl#4-#5) (Fig. 2b-c; Extended Data Fig. 2b). Unspecific binding to the 120
used beads and/or the anti-NCAM-L1 antibody were excluded using an immunoblot 121
approach (Extended Data Fig. 2a). Comparing NCAM-L1 levels of NEs of PD patients and 122
controls showed no significant differences (Fig. 2c). Next, different isoforms of NCAM-L1 123
were detected through western blot analyses in samples containing NEs (Ctrl#1-#2, PD#3) 124
(Fig. 2d, Extended Data Fig. 2d). Unspecific binding to the anti-NCAM-L1 antibody and/or the 125
anti-NCAM-L1 antibody with beads were also excluded through western blot analysis 126
(Extended Data Fig. 2c). For both approaches the protocol of NEs-isolation was performed 127
as usual. A significant increase in NCAM-L1 levels was detected for NEs samples (Fig. 2e) 128
and the purification of NEs resulted in a significant increase of other established neuronal 129
markers as synaptophysin
24
, the pan-neuronal marker protein gene product 9.5 (PGP9.5, 130
also known as ubiquitin C-terminal hydrolase L1 (UCHL-1)) and neuron-specific enolase 131
(NSE)
25,26
, which confirm the homogenous and neuronal origin (Fig. 2f; Extended Data 132