Patients with isolated REM-sleep behavior disorder have elevated levels of alpha-synuclein aggregates in stool

Abstract

Misfolded and aggregated α-synuclein is a neuropathological hallmark of Parkinson’s disease (PD). Thus, α-synuclein aggregates are regarded as a biomarker for the development of diagnostic assays. Quantification of α-synuclein aggregates in body fluids is challenging, and requires highly sensitive and specific assays. Recent studies suggest that α-synuclein aggregates may be shed into stool. We used surface-based fluorescence intensity distribution analysis (sFIDA) to detect and quantify single particles of α-synuclein aggregates in stool of 94 PD patients, 72 isolated rapid eye movement sleep behavior disorder (iRBD) patients, and 51 healthy controls. We measured significantly elevated concentrations of α-synuclein aggregates in stool of iRBD patients versus those of controls (p = 0.024) or PD patients (p < 0.001). Our results show that α-synuclein aggregates are excreted in stool and can be measured using the sFIDA assay, which could support the diagnosis of prodromal synucleinopathies.

Introduction

Parkinson’s disease (PD), dementia with Lewy bodies (DLB), and multiple system atrophy (MSA) are synucleinopathies characterized by the oligomerization and aggregation of α-synuclein into assemblies rich in cross-beta structure, which is a critical event in the pathogenesis of these diseases^1,2,3,4,5,6. Because α-synuclein fibrils deposit in Lewy bodies and Lewy neurites in neurons in PD and DLB or in neuronal and glial cytoplasmic inclusions in MSA, small oligomeric intermediates of α-synuclein that are still soluble are thought to be the most toxic to neurons^7,8,9. The diagnosis of synucleinopathies primarily relies on clinical assessment and brain imaging, which are either imprecise or costly, especially during early disease stages, highlighting the urgent need for a biomarker for early and reliable disease detection^10,11. Also, for potential therapies to be most effective, detecting synucleinopathies at early stages may be advantageous when the burden of pathologic α-synuclein and neuronal loss are minimal. In the prodromal stage of synucleinopathies, non-motor symptoms, such as hyposmia and gastrointestinal dysfunction, can appear as early symptoms up to 20 years before cardinal motor symptoms manifest^10,12. Also, most patients with isolated rapid eye movement (REM) sleep behavior disorder (iRBD), a parasomnia characterized by REM sleep without atonia and vivid dream enactment, convert to PD, DLB, or MSA within 10–20 years of diagnosis^13,14. IRBD patients display many non-motor symptoms seen in PD patients including hyposmia, orthostatic hypotension, and gastrointestinal dysfunction¹⁵. Importantly, iRBD patients also accumulate pathologic α-synuclein assemblies in the central and peripheral nervous system^16,17. Besides the nervous system and cerebrospinal fluid (CSF), α-synuclein aggregates have been detected in skin, olfactory mucosa, saliva, tears, urine, and blood of PD patients, which may all serve as potential sources for diagnosing synucleinopathies^{18,19,20,21,22,23,24,25,26}. Histological findings also show that enteric neurons in the gastrointestinal submucosa of PD patients contain pathologic α-synuclein, even at prodromal stages^27,28,29. Together with findings of pathological α-synuclein in the enteric nervous system (ENS) and the dorsal motor nucleus of the vagus nerve and the anterior olfactory bulb as the earliest lesion sites in PD, Braak and colleagues postulated the dual-hit hypothesis suggesting that α-synuclein pathology spreads from the olfactory bulb to the temporal lobes, and from the ENS via the vagus nerve to the CNS^30,31. Further, several studies suggest that pathologic α-synuclein assemblies have prion-like properties, allowing them to multiply and propagate between neurons in the nervous system and to spread across large distances in the brain and body, including from the ENS to the CNS and vice versa^32,33,34,35.

Earlier findings in transgenic TgM83^+/− mice expressing human α-synuclein demonstrated that α-synuclein fibrils used for an oral challenge could cross the mucosal barrier of the gastrointestinal tract after which they invaded the nervous system and triggered a synucleinopathy in these mice³⁶. Here, we hypothesized that human pathologic α-synuclein assemblies released from, e.g., affected enteric neurons may traverse the mucosal lining of the gastrointestinal tract and be excreted in stool. This would involve a mechanism similar to the environmental shedding of prions in feces of deer and elk infected with chronic wasting disease or that of sheep and goats infected with scrapie^37,38.

It remains challenging to specifically detect protein aggregates in body fluids as the expected concentrations are very low, requiring a high sensitivity that standard techniques such as ELISA do not provide³⁹. Moreover, monomers are present in great excess, necessitating that assays have a high selectivity for oligomers. To overcome these challenges in detecting aggregated α-synuclein assemblies, we used the sFIDA (surface-based fluorescence intensity distribution analysis) assay, which uses antibodies targeting overlapping or identical epitopes, enabling it to specifically detect aggregates in the presence of monomers^40,41,42,43. Here, the Syn211 antibody that recognizes amino acids 121–125 on α-synuclein is attached to a glass surface and captures α-synuclein species in stool homogenates⁴⁴. Next, the captured α-synuclein species are detected with a blend of two Syn211 antibodies, each labeled with a different fluorescent dye, one green and one red (Fig. 1). The surface is then scanned by two-channel fluorescent confocal microscopy, yielding single particle sensitivity. Because the Syn211 antibody only recognizes a single linear epitope on α-synuclein, using the same Syn211 antibody to capture and detect α-synuclein species guarantees specific quantitation of only aggregated α-synuclein species. Monomers with a single linear epitope are captured but cannot be detected and are thus disregarded. Since the sFIDA assay in its current version does not differentiate between α-synuclein species from small oligomers to large fibrils, we refer to analytes measured in sFIDA as aggregates.

We recently used sFIDA to quantitate α-synuclein and tau aggregates in CSF of patients with PD, DLB, and other neurodegenerative diseases, showing that sFIDA is a reliable, highly sensitive, and robust assay for the diagnosis of neurodegenerative diseases and drug development⁴⁰. The aim of this study was to detect and quantify α-synuclein aggregates in human stool and, more importantly, to demonstrate its possible use in a clinical setting and potentially for drug development. We adapted the sFIDA assay to quantify α-synuclein aggregates in stool samples of 217 subjects. We technically validated this assay by assessing parameters like the limit of detection (LOD), coefficient of variation (CV%), inter-assay correlation, and cross-reactivity to other aggregates.

Results

Descriptive analysis of the patient and control groups

We collected stool samples of PD (n = 94) and iRBD patients (n = 72) and healthy controls (HC) without past or present neurological disorder (n = 51) to determine if and how much α-synuclein aggregates they contain. Demographic and clinical information for these three cohorts on sex, age, education, disease duration, memory, motor, non-motor, olfactory, constipation, and further information are available in Table 1, and for individual patients and healthy controls in Supplementary Table 1. There was a gender bias towards men in the PD (69.1%) and more so in the iRBD (86.1%) cohort. In the control group, gender bias was slightly more towards women (60.8%). The age of PD (64.5 ± 9.9 years) and iRBD patients (66.3 ± 6.4 years) did not significantly differ, while healthy controls were on average 8–10 years younger (56.4 ± 17.1 years). We did not observe significant differences in the duration of education between the three groups. Disease duration in PD patients was 9.0 ± 5.8 years and in iRBD patients 7.4 ± 5.5 years. The cognitive performance (DemTect score) of healthy controls (15.9 ± 2.3) was significantly higher than that of PD (13.8 ± 3.3) and iRBD patients (14.8 ± 2.3, Supplementary Fig. 1a)⁴⁵. PD patients (24.1 ± 14.9) scored significantly higher than iRBD patients (4.0 ± 2.7) on the Movement Disorder Society’s Unified Parkinson’s Disease Rating Scale Part III (MDS-UPDRS III)⁴⁶. Equally, PD patients (30.6 ± 23.1) scored significantly higher on the Non-Motor Symptoms Scale (NMSS) for PD than iRBD patients (16.2 ± 18.0) or healthy controls (12.5 ± 14.4, Supplementary Fig. 1b)⁴⁷. Also, PD patients (4.1 ± 4.0) were significantly more constipated than iRBD patients (2.8 ± 2.7) and healthy controls (2.5 ± 2.5), based on the Cleveland Clinic Constipation Scoring System (Supplementary Fig. 1c)⁴⁸. PD patients (5.3 ± 2.6) also suffered more from hyposmia (Supplementary Fig. 1d), reflected in a significantly lower scent score, than iRBD patients (6.6 ± 2.7) or healthy controls (10.4 ± 1.7). In addition, PD patients (5.6 ± 2.2) scored significantly higher in the screening questionnaire for parkinsonism compared to iRBD patients (0.3 ± 0.8) and healthy controls (0.3 ± 0.8, Supplementary Fig. 1e)⁴⁹. The results of the RBD screening questionnaire (RBDSQ, Supplementary Fig. 1f) also showed significant differences between iRBD (9.0 ± 2.7) and PD patients (4.9 ± 3.0) or healthy controls (1.7 ± 1.6), as well as between PD patients and healthy controls⁵⁰. On average, PD patients scored 3.0 ± 0.9 on the Hoehn and Yahr scale and 40 ± 20% on the levodopa challenge test⁵¹.