Clinical Validation of α‐Synuclein in Salivary Extracellular Vesicles as a Biomarker for Parkinson's Disease: A Longitudinal Study

ABSTRACT

Background

Parkinson's disease (PD) is a progressive neurodegenerative disorder characterized by the accumulation of misfolded alpha‐synuclein (α‐Syn) and the subsequent loss of dopaminergic neurons. Identifying reliable and non‐invasive biomarkers is crucial for accelerating early diagnosis and monitoring disease progression. To this aim, we longitudinally investigated α‐Syn in salivary extracellular vesicles (SEVs) in PD patients and the correlation with clinical outcomes.

Methods

SEVs were isolated from PD patients and healthy controls (HCs) saliva using differential ultracentrifugation followed by morphological and molecular characterization. The levels of both total (α‐Syn_Tot) and oligomeric (α‐Syn_Olig) α‐Syn were quantified by ELISA.

Results

We found a significant increase in both α‐Syn_Tot and α‐Syn_Olig in PD‐derived SEVs compared to HCs, and receiver operating characteristic analysis revealed that α‐Syn_Olig displayed higher sensitivity (65%) for discriminating PD from HCs compared to α‐Syn_Tot (59%). Moreover, α‐Syn_Olig levels correlated negatively with Mini‐Mental State Examination scores and were higher in patients with motor fluctuations. Finally, we found that α‐Syn_Olig levels did not change after one‐year follow‐up in patients when also the clinical parameters remained unaltered.

Conclusions

These results establish for the first time that SEVs‐associated α‐Syn_Olig is a promising, sensitive and non‐invasive biomarker for PD diagnosis and clinical correlation studies, bearing higher sensitivity than α‐Syn_Tot. Moreover, α‐Syn_Olig levels closely followed the clinical outcomes in PD patients. Finally, these findings strengthen the rationale for the further exploration of SEVs to disclose still unavailable accessible biomarkers for multiple neurological diseases.

Salivary Extracellular vesicles (SEVs) isolated from PD patients exhibit increased levels of both total (α‐Syn_Tot) and oligomeric alpha‐synuclein (α‐Syn_Olig). In PD patients, SEVS‐α‐Syn_Olig levels correlate with cognitive decline (MMSE) and motor fluctuations (MDS‐UPDRS IV) and remain stable after 1 year follow‐up. These findings identify SEVs‐α‐Syn_Olig as a sensitive, non‐invasive biomarker reflecting key clinical features of Parkinson's disease.

1. Introduction

Parkinson's disease (PD) is the second most common progressive neurodegenerative disorder, characterized by heterogeneous clinical manifestations, including motor and non‐motor symptoms that significantly impact patient quality of life [1]. As with other synucleinopathies, PD pathology involves the aggregation of misfolded alpha‐synuclein (α‐Syn) into cytoplasmic inclusions known as Lewy Bodies (LBs), which leads to dopaminergic neuronal loss in the brain [2]. Importantly, motor symptoms appear only after substantial neuronal damage, with up to 80% of nigrostriatal neurons compromised by the time of clinical diagnosis [3, 4]. This delayed symptom onset underscores the urgent need for biomarkers to facilitate early diagnosis, monitor disease progression, and assess therapeutic efficacy in clinical trials [5].

Due to its central role in disease pathology, the evaluation of α‐Syn as a prognostic biomarker for PD is currently under close investigation [5, 6]. Interestingly, α‐Syn can cross the blood–brain barrier and is detectable in peripheral biofluids such as cerebrospinal fluid (CSF), plasma, urine, and saliva [7]. Under physiological conditions, α‐Syn exists in a monomeric form (α‐Syn_Tot) involved in neurotransmitter release. However, monomers can misfold and aggregate into oligomers (α‐Syn_Olig) within LBs, which in turn can drive the appearance of alterations like impaired neurotransmission and neurodegeneration [8].

One proposed pathogenic mechanism for PD is based on the release and spread of α‐Syn in the brain via extracellular vesicles (EVs), membrane‐enclosed nanoparticles secreted virtually by all cell types, including neurons [9, 10]. EVs contain a variety of molecules, such as proteins and nucleic acids (e.g., mRNAs and miRNAs), that can be released either to remove excess or toxic cellular content or transported and captured by neighboring cells within the process of long‐range cell‐to‐cell communication. Intriguingly, mounting evidence suggests that EVs play a crucial role in PD progression by transferring α‐Syn_Olig from damaged neurons to healthy ones, potentially promoting neurodegeneration [10, 11, 12]. On the other hand, because α‐Syn carried within EVs exhibits greater stability than its free form in biofluids, possibly due to the protection exerted by the EVs' lipid bilayer [13], it is under close investigation as a putative biomarker for PD [14].

Although several studies focused on isolating circulating EVs from CSF and plasma, more recently the interest of using salivary EVs (SEVs) to assess α‐Syn levels has emerged [15, 16, 17]. Indeed, saliva is readily accessible, its collection is non‐invasive and does not require clinical settings; therefore, minimizing compliance challenges and costs. These characteristics make SEVs a feasible source for biomarkers discovery across a variety of conditions, including cancer, cardiovascular, autoimmune, and neurodegenerative diseases [18]. Because the establishment of reliable biomarkers for PD is a critical medical need, in this study we validated the methodology for obtaining high‐yield isolation of SEVs in a longitudinal evaluation of a cohort of PD patients in order to validate the use of α‐Syn loaded into SEVs as diagnostic and/or prognostic biomarker for PD progression.

2. Materials and Methods

The reported cross‐sectional and longitudinal study was approved and supervised by the Ethical Committee of the University of Turin (N° 0040679). Informed consent was obtained from the participants. A total of 100 subjects (63 PD patients and 37 healthy controls – HCs) were recruited from the Movement Disorder Clinic of the Department of Neuroscience of the University of Turin. Inclusion criteria were clinical PD diagnosis based on Movement Disorder Society (MDS) Diagnostic Criteria [19]. Exclusion criteria: diagnosis of other neurological, psychiatric or medical conditions and drug‐induced parkinsonism. Subjects were also excluded from the study if affected by chronic inflammatory, autoimmune or cardiovascular diseases, diabetes mellitus, neoplasms, and salivary gland/oral cavity pathologies. HCs were age‐matched subjects accompanying the patient, mostly partners or caregivers and were excluded if they were affected by neurological or psychiatric disorders. Twenty PD patients were prospectively evaluated after a 1‐year period for clinical parameters and saliva collection.

2.1. Clinical Assessment

Clinical assessment at baseline of PD patients included collection of demographic characteristics and clinical features. For each patient we collected age at onset, disease duration, and Levodopa equivalent daily dose (LEDD). Motor impairment was assessed using the MDS–Unified Parkinson's Disease Rating Scale (UPDRS) part III and selected items from part II (items 2.10, 2.12, 2.13). The tremor dominant (TD) score was calculated as the sum of tremor‐related items, while the postural instability and gait disturbance (PIGD) score was obtained by summing the corresponding items. Subsequently, patients were classified into PIGD and non‐PIGD phenotypes based on the TD/PIGD ratio: PD patients with a TD/PIGD ratio ≤ 0.90 were assigned to the PIGD group, whereas those with a ratio > 0.90 were categorized as non‐PIGD [20, 21, 22]. Motor fluctuation was defined by UPDRS part IV and motor severity by Hoehn and Yahr (H&Y) staging system. The presence of dysautonomia was confirmed if patients had at least one symptom or positive diagnostic test (ambulatory blood pressure monitoring or cardiovascular reflexing tests). Cognitive function was evaluated with corrected Mini‐Mental State Examination (MMSE) and threshold for cognitive impairment was < 26.

2.2. Salivary Sample Collection

Saliva was collected during the participant visit (8:00–12:00 am), in a resting state, by drooling into 50 mL falcon tubes a comparable amount of saliva between patients and HCs. All subjects refrained from eating, drinking, smoking, or oral hygiene for at least 1 h prior to sample collection. Besides, 5 min before collection, the participants rinsed their mouths with water for 1 min to remove any tissue and debris [15].

2.3. Isolation and Characterization of SEVs

EVs were isolated from both PD patients and HCs saliva samples by differential ultracentrifugation [23]. 24 h after collection, samples were diluted (1:1) in sterile‐filtered 0.01 M‐PBS pH 7.4, centrifuged at 3000 g for 30 min at 4°C to remove particles (e.g., bacteria, dead cells, cell debris). The supernatant was filtrated using 0.22 μm filters into an ultracentrifuge tube previously placed on ice and then ultracentrifuged at 100,000 g for 2 h at 4°C (Optima L‐60, 50.2 Ti rotor, Beckman Coulter) to pellet EVs [23]. Subsequently, the pellet was washed with sterile‐filtered 0.01 M‐PBS and ultracentrifuged again at 100,000 g for 2 h at 4°C to remove contaminating proteins. The resulting EVs‐containing pellets were resuspended in: (1) 1% DMSO‐PBS for both Nanoparticle Tracking Analysis (NTA) and transmission electron microscopy (TEM); (2) ice‐cold lysis RIPA buffer (150 mM NaCl, 1% Triton X‐100, 0.5% Na‐deoxycholate, 0.1% Na‐dodecyl‐sulfate, 50 mM Tris‐Base, 5 mM EDTA, 1 mM EGTA) containing protease and phosphatase inhibitors (100 mM DTT, 1 M NaF, 200 mM NA‐orthovanadate, 100 mM PMSF) for both western blot (WB) and enzyme‐linked immunosorbent assay (ELISA). All resuspended samples were stored at −80°C until their use.

SEVs concentration and size were measured using a NanoSight LM10 system (NanoSight, Salisbury, UK). Briefly, the Brownian movement of the SEVs was detected using a 405 nm laser light source, videorecorded and analyzed using the NTA 3.1 Software. The software analyzed three 30‐s videos recorded with camera level set at 14 [24], threshold value was set to detect between 15 and 100 particles per frame. The same settings were maintained throughout the analyses of all samples.

For TEM inspection, a 20 μL drop of PBS‐resuspended SEVs was placed on 200 mesh nickel formvar carbon‐coated grids (Electron Microscopy Science, Hatfield, USA), let to adhere (dry) for 5 min before incubation with 2.5% glutaraldehyde (5 min). Grids were then washed with distilled water, negatively stained using Nano‐W and NanoVan solutions (Nanoprobes, Yaphank, NY, USA), and imaged using a JEM‐1400 Flash transmission electron microscope (JEOL, Tokyo, Japan) [24].

WB was performed following a standard protocol [25]. SEVs lysates were boiled in SDS sample buffer and separated using a 10% SDS‐PAGE gel. Proteins were transferred to PVDF membranes, subsequently blocked in 5% non‐fat milk in TBS (20 mM Tris–HCl, pH 7.5, 150 mM NaCl) + 0.1% Tween‐20 (TBS‐T) for 1 h and then incubated with primary antibodies overnight at 4°C. After several washes with TBS‐T, membranes were incubated with appropriate secondary antibodies for 1 h at room‐temperature (Table 1). The chemiluminescent signal was visualized using Westar Antares ECL Blotting Substrates (Cyanagen; Italy), acquired with Bio‐Rad ChemiDocTM Imagers (Bio‐Rad; Italy) and the optical density (O.D.) values were analyzed with Image‐J software (NIH, USA).

TABLE 1.

Antibodies used.

	Species of origin	WB	Supplier and catalog no.
Primary antibody
CD63	Rabbit	1:1000	Immunological Sciences, Italy, #AB‐83481
CD9	Rabbit	1:500	Immunological Sciences, Italy, #AB‐83880
β‐tubulin	Mouse	1:500	Immunological Sciences, Italy, #MAB‐80143
Calnexin	Rabbit	1:250	Cell signaling, USA, #2433S
TGS101	Mouse	1:200	Santa Cruz Biotech., USA, #sc‐7964
Alix	Mouse	1:200	Santa Cruz Biotech., USA #sc‐53,540
Secondary antibodies
Anti‐rabbit		1:5000	Sigma, Italy, #A0545
Anti‐mouse		1:5000	Sigma, Italy, #A4416

2.4. α‐Syn ELISA

The concentration of SEVs‐α‐Syn_Tot was determined with an anti‐α‐Syn quantitative ELISA Kit (SensoLyte, Anaspec, USA), while for SEVs‐α‐Syn_Olig a human α‐Syn oligomer ELISA Kit (MyBioSource, USA) was used. O.D. was determined at 450 nm using an iMark Bio‐Rad microplate reader (Bio‐Rad; Italy). The kit protocols were followed to build the standard curve and quantify protein levels. Each run included PD, HC samples, and blank controls used as normalizer. α‐Syn values were normalized to SEV concentration in each sample.

2.5. Statistical Analysis

Prism software (GraphPad, La Jolla, USA) was used to perform unpaired Student's t‐tests to compare the concentrations of both α‐Syn_Tot and α‐Syn_Olig in PD patients and HCs. The area under (AUC) of the receiver operating characteristic (ROC) curve was calculated to determine both sensitivity and specificity of the ELISA tests. The Spearman and Pearson's correlation test were used to correlate clinical data, including age, age at disease onset, disease duration, H&Y, MDS‐UPDRS (part III and IV), MMSE, and TD/PIGD with the two α‐Syn isoform expressions. A difference with a p‐value < 0.05 was considered statistically significant.

Because of technical issues that occurred during either filtration and ultracentrifugation of SEVs, the following number of samples were omitted from further analyses: 13 out of 100 and 4 out of 20 for the 1‐year follow‐up. Moreover, the Modified Thompson Tau‐Test identified 20 outliers in the analysis of SEVs concentration (7) and in the ELISA assessment (13).

3. Results

3.1. Participant's Demographic and Clinical Features

For the baseline investigation, saliva samples were collected from 63 PD patients and 37 cross‐matched aged HCs. After the procedures for SEVs isolation and analytical processing, the cohort was reduced to 56 PD patients and 30 HCs due to the application of exclusions criteria (see Section 2). Demographic and clinical characteristics of participants are listed in Table 2. Of all recruited patients diagnosed with idiopathic PD, only one was genetically tested and exhibited a LRRK2 mutation.

TABLE 2.

Demographic and clinical data.

	HCs (30)	PD (57)
Demographic parameters
Age (years) (mean ± SD)	59.10 ± 12.74	60.37 ± 11.01
Sex
Male/Female	8/22	41/16
Neurological parameters
Disease onset, years (mean ± SD)	—	50.67 ± 9.1
Disease duration, years (mean ± SD)	—	9.67 ± 7.17
MDS‐UPDRS part III (mean ± SD)	—	27.91 ± 12.59
MDS‐UPDRS part IV (mean ± SD)	—	2.47 ± 3.72
H&Y stage (median, min–max)	—	2.1 (1–4)
1 (n; %)	—	15; 26.31%
2 (n; %)	—	26; 45.61%
3 (n; %)	—	12; 21.05%
4 (n; %)	—	4; 7.03%
MMSE (mean ± SD)	—	27.32 ± 3.2
Dysautonomia (n; %)	—	23; 40.35%
PIGD score (mean ± SD)	—	3.60 ± 3.73
Phenotypes
Non‐PIGD (n; %)		34; 64.15%
PIGD (n; %)		19; 35.85%
TD score (mean ± SD)	—	4.94 ± 4.24
LEDD, mg/d (mean ± SD)	—	687 ± 540.6
DBS (n; %)	—	7; 13.73%
Gene characterization
LRRK2 (n; %)	—	1; 2.78%

Abbreviations: DBS, Deep Brain Stimulation; HCs, healthy controls; H&Y, Hoehn and Yahr Scale (stage); LEDD, levodopa equivalent daily dose; MDS‐UPDRS part III, Movement Disorder Society‐Unified Parkinson's Disease Rating Scale‐Part‐III; MDS‐UPDRS part IV, Movement Disorder Society‐Unified Parkinson's Disease Rating Scale‐Part‐IV; MMSE, Mini‐Mental State Examination; PD, Parkinson's Disease; PIGD, postural instability gait difficulty; TD, tremor dominant.

3.2. Isolation and Characterization of SEVs

We characterized SEVs by analyzing the expression of selective EV markers: tetraspanins (i.e., CD63 and CD9), as well as Alix and TSG101, integrated components of the ESCRT (Endosomal Sorting Complex Required for Transport) cellular machinery essential for EVs biogenesis, following the International Society for Extracellular Vesicles “Minimal Information for Studies of Extracellular Vesicles” (MISEV) guidelines [26]. WB analysis revealed that all these markers were expressed in SEVs from both PD patients and HCs. In contrast, the immunosignal for calnexin, a marker of the endoplasmic reticulum [27] that signals poor EVs purity, was virtually absent from isolated SEVs. Intriguingly, we also found that β‐tubulin, a marker of neuronal‐derived EVs, was expressed in SEVs from both experimental groups (Figure 1A). Moreover, TEM inspection revealed that SEVs show the canonical round cup‐shaped morphology with a preserved membrane (Figure 1B), while NTA, employed to assess the size distribution of SEVs, revealed a broad peak centred around 200 nm (Figure 1C). These data indicate that the average size of the isolated SEVs is consistent with that of the microvesicles population [28]. Finally, we found that while SEVs average size did not differ between the two groups (PD: 204.4 ± 4.57; HC: 202.5 ± 6.09; p > 0.05; Figure 1D), the concentration of SEVs in PD samples was greatly reduced compared to HCs (PD: 5.1 ± 0.5 × 10¹⁰; HC: 7 ± 0.9 × 10¹⁰; *p < 0.05; Figure 1E).

FIGURE 1.

Characterization of SEVs. (A) Western blotting showing the expression of positive (Alix, CD63, TSG101, CD9, and β‐tubulin) and negative (calnexin) markers in SEVs lysates from HCs, PD patients, and in mouse brain total lysate used as control (Ctrl). (B) Transmission electron microscopy images of SEVs. (C–E) Analysis of size distribution determined by nanoparticle tracking analysis (C) showing the size (D) and the concentration (E) of SEVs from PD patients and HCs. HC, Healthy control, PD, PD patient. Scale bar: 100 nm.

3.3. Quantification of α‐Syn in SEVs Cargo by ELISA

We next analyzed the level of both total (α‐Syn_Tot) and oligomeric (α‐Syn_Olig) forms of α‐Syn contained in SEVs with commercially available ELISA kits, a technique routinely used in clinical analyses. Intriguingly, α‐Syn_Tot levels were found significantly higher in PD patients compared to HCs (PD: 2.75 ± 0.35 pg/mL; HCs: 1.45 ± 0.24 pg/mL; *p < 0.05; Figure 2A). A similar pattern was observed for α‐Syn_Olig levels, which were also elevated in PD patients (PD: 6.77 ± 0.59 ng/mL; HCs: 4.2 ± 0.76 ng/mL; *p < 0.05; Figure 2B). Notably, the patient carrying the LRRK2 mutation exhibited lower α‐Syn_Olig levels (0.57 ng/mL), consistent with previous findings [29]. To assess the diagnostic accuracy of α‐Syn_Tot and α‐Syn_Olig levels in SEVs, we performed a ROC analysis (Figure 2C). As shown in Table 3 and Figure 2C, when cut‐off values of 1.64 pg/mL and 4.76 ng/mL were applied for α‐Syn_Tot and α‐Syn_Olig, respectively, the sensitivity and specificity were 59% and 74% for α‐Syn_Tot, and 65% and 75% for α‐Syn_Olig. Additionally, the AUC was 0.68 for α‐Syn_Tot and 0.72 for α‐Syn_Olig, values that indicate an overall fair diagnostic accuracy of the test.

FIGURE 2.

α‐Syn load in SEVs is different in PD patients. (A, B) Bar graphs showing the mean concentration of α‐Syn_Tot (A) and α‐Syn_Olig (B) carried by SEVs from PD patients and HCs. The analysis revealed a higher level of both α‐Syn_Tot and α‐Syn_Olig in PD patients' EVs compared to HCs. (C) ROC curve analysis provided an AUC of 0.68 for α‐Syn_Total (p < 0.05) and an AUC of 0.71 for α‐Syn_Olig (p < 0.05). Student's t‐test, *p < 0.05. ROC: Receiver operating characteristic, AUC: Area Under the Curve.

TABLE 3.

ROC curve analysis of α‐Syn_Tot and α‐Syn_Olig levels in SEVs between PD and HCs.

	AUC	p	Cut‐off	Sensitivity	Specificity
α‐SynTot, pg/mL	0.68	0.017*	1.635	59%	74%
α‐synOligo, ng/mL	0.71	0.004*	4.756	65%	75%

Note: * indicates p < 0.05.

Abbreviation: AUC, area under the curve.

3.4. Correlational Analyses Between α‐Syn Content and Clinical Parameters

To investigate whether the levels of α‐Syn isoforms in SEVs could be used as indicators of disease severity and progression, we stratified the PD cohort based on clinical characteristics, including the presence of dysautonomia, motor fluctuations (MDS‐UPDRS IV ≥ 1), cognitive deficits (MMSE < 26), H&Y stages, and PIGD phenotype. We found that α‐Syn_Olig concentration was significantly higher in patients with motor fluctuation (MDS‐UPDRS part IV ≥ 1) and with cognitive impairment (MMSE < 26). Moreover, we observed a trend toward increased α‐Syn_Olig levels in PIGD patients (TD/PIGD ratio ≤ 0.90; Table 4). Pearson's and Spearman's correlation analyses revealed a negative, significant correlation between α‐Syn_Olig levels and MMSE scores (r = −0.36; *p < 0.05), and a positive correlation between α‐Syn_Olig levels and MDS‐UPDRS part III which did not reach the threshold for statistical significance (r = 0.28; p = 0.05) as did the positive correlation between α‐Syn_Tot levels and TD/PIGD ratio (r = 0.34; p = 0.06). Moreover, a modest, not‐significant positive correlation was observed between levodopa equivalent daily dose (LEDD) and α‐Syn_Olig levels (r = 0.24; p = 0.09), while α‐Syn_Tot levels showed a moderate negative, not statistically significant, correlation with LEDD (r = −0.29; p = 0.07). The entire dataset of the correlation studies is reported in Table 5.

TABLE 4.

Correlations analyses of α‐Syn_Tot and α‐Syn_Olig concentration in SEVs with PD patients clinical scores.

	α‐SynTot, pg/mL	p	α‐SynOligo, ng/mL	p
Dysautonomia
No	2.74 ± 0.46	0.98	6.38 ± 0.81	0.43
Yes	2.76 ± 0.53		7.33 ± 0.82
MMSE
< 26	2.43 ± 0.57	0.74	8.83 ± 1.15	0.03*
≥ 26	2.71 ± 0.40		5.87 ± 0.68
Motor fluctuation
No (MDS‐UPDRS part IV 0)	3.02 ± 0.46	0.37	5.33 ± 0.56	0.003**
Yes (MDS‐UPDRS part IV ≥ 1)	2.37 ± 0.52		8.68 ± 1.02
Phenotype
Non‐PIGD	2.96 ± 0.58	0.17	4.97 ± 0.46	0.07
PIGD	1.92 ± 0.41		7.02 ± 0.99
MDS‐UPDRS part III
Mild < 32	2.49 ± 0.46	0.31	6.32 ± 0.63	0.28
Moderate/severe > 33	3.25 ± 0.49		7.68 ± 1.23
H&Y stage
Mild/moderate 1–2	2.78 ± 0.40	0.89	7.09 ± 0.68	0.39
Severe 3–4	2.67 ± 0.70		5.96 ± 1.16

Note: Data are presented as mean ± SEM. * indicates p < 0.05, ** indicates p < 0.005.

Abbreviations: H&Y, Hoehn and Yahr Scale (stage); MDS‐UPDRS‐III and IV, Movement Disorder Society‐Unified Parkinson's Disease Rating Scale‐Part‐III and IV; MMSE, Mini‐Mental State Examination; PIGD, postural instability and gait difficulty.

TABLE 5.

Correlation analyses of SEVs α‐Syn_Tot and α‐Syn_Olig concentrations with clinical parameters.

	Correlation coefficient (r) α‐SynTot	p	Correlation coefficient (r) α‐SynOligo	p
Demographic & neurological parameters
Age	−0.19	0.23	0.11	0.44
Disease onset	−0.11	0.50	0.04	0.77
Disease duration	−0.14	0.38	0.18	0.23
MDS‐UPDRS part III	0.03	0.83	0.28	0.05
H&Y stage	−0.07	0.65	0.01	0.93
TD/PIGD ratio	0.32	0.10	−0.26	0.10
MMSE	0.17	0.30	−0.36	0.02*
LEDD	−0.29	0.07	0.24	0.09

Note: * indicates p < 0.05.

Abbreviations: H&Y, Hoehn and Yahr Scale (stage); LEDD, L‐dopa equivalent daily dose; MDS‐UPDRS‐III, Movement Disorder Society‐Unified Parkinson's Disease Rating Scale‐Part‐III; MMSE, Mini‐Mental State Examination; PIGD, postural instability and gait disturbance; TD, tremor dominant.

3.5. Longitudinal Evaluation of SEVs α‐Syn_Olig and PD Progression

The one‐year follow‐up clinical assessment of 20 PD patients revealed no significant changes in their neurological parameters (e.g., UPDRS III‐IV, H&Y stage, and MMSE; Table 6). Given that our ROC analysis results indicate that α‐Syn_Olig outperforms α‐Syn_Tot in diagnostic accuracy because of higher sensitivity (65% vs. 59%; Table 3), we prioritized the analysis of this biomarker in the follow‐up study. Interestingly, the analyses of α‐Syn_Olig levels in SEVs of the 20 PD follow‐up patients did not show any significant difference after 1 year (Table 6), and no differences in both size and concentration of SEVs were detected. Thus, these data indicate that, at least after 1 year, the absence of changes in clinical parameters is reflected by unaltered levels of SEVs α‐Syn_Olig levels in PD patients.

TABLE 6.

Longitudinal examination of PD patients cohort.

	PD baseline	PD follow up	p
Demographic parameters
Age (years) (mean ± SD)	61.2 ± 10.43	62.4 ± 10.14	0.71
Neurological parameters
MDS‐UPDRS III (mean ± SD)	23.55 ± 12.82	22.70 ± 13.35	0.84
MDS‐UPDRS IV (mean ± SD)	3.5 ± 4.73	3.45 ± 3.50	0.97
H&Y stage (median)
1 (n; %)	5; 25%	5; 25%
2 (n; %)	11; 55%	13; 65%
3 (n; %)	4; 20%	2; 10%
4 (n; %)	—	—
MMSE (mean ± SD)	27.27 ± 2.83	27.69 ± 3.54	0.44
Dysautonomia (n; %)	8; 40%	7; 35%
PIGD score (mean ± SD)	0.7 ± 0.68	0.72 ± 0.59	0.91
LEDD, mg/d (mean ± SD)	637.1 ± 478.6	704.2 ± 443.3	0.65
DBS (n; %)	5; 25%	7; 35%
SEVs examination
SEVs concentration, particles/mL (mean ± SEM)	5.68E+10 ± 6.30E+10	5.69E+10 ± 2.48E+10	0.99
SEVs size, nm (mean ± SEM)	203.6 ± 6.97	213.9 ± 6.48	0.3
SEVs α‐SynOligo, ng/mL (mean ± SEM)	5.94 ± 0.65	5.52 ± 0.9	0.3

Abbreviations: DBS, Deep Brain Stimulation; H&Y, Hoehn and Yahr Scale (stage); LEDD, levodopa equivalent daily dose; MDS‐UPDRS‐III and IV, Movement Disorder Society‐Unified Parkinson's Disease Rating Scale‐Part‐III and IV; MMSE, Mini‐Mental State Examination; PD, Parkinson's disease; PIGD, postural instability and gait disturbance; SEVs, salivary extracellular vesicles; TD, tremor dominant.

4. Discussion

Despite significant progress in understanding the neuropathogenesis of PD, early diagnosis remains challenging due to its poor accuracy. To address this issue, it is crucial to identify biomarkers that are reliable, easily accessible, and highly sensitive to enhance diagnostic precision. Thus, in this study, we assessed the feasibility of setting up a pipeline, reproducible in standard clinical settings, to isolate SEVs and quantify the α‐Syn load, to establish a non‐invasive platform for biomarkers with potential diagnostic and prognostic applications for PD. The main finding of this study is that both α‐Syn_Tot and α‐Syn_Olig are robustly increased in SEVs from PD patients compared to HCs. Moreover, the approach used here reveals that non‐invasive SEVs collection offers a practical advantage for patient compliance and repeated sampling, while suggesting focusing in future studies on α‐Syn_Olig as a biomarker with stronger diagnostic and prognostic potential highlights the importance of this research. Additionally, the results of the longitudinal observation indicate that a one‐year follow‐up interval is insufficient to detect significant changes in α‐Syn_Olig levels in SEVs, underlining that extended observations should be considered when planning future studies to capture disease progression.

The accumulation of α‐Syn aggregates in the brain is the most recognized characteristic of a group of neurodegenerative diseases, including PD. Because α‐Syn levels are increased in EVs isolated from plasma/serum and CSF of PD patients [30], several studies investigated α‐Syn content in EVs as an early diagnostic biomarker for this disease [6]. However, the invasive nature of CSF collection and the risk of blood contamination in serum/plasma preparations introduce significant limitations to these approaches, making SEVs‐based investigation a promising non‐invasive alternative [15]. Interestingly, several studies have shown that α‐Syn was found in nerve fibers innervating salivary glands [31, 32, 33, 34]. Similarly, submandibular gland biopsies from PD patients revealed positive staining for α‐Syn [32], further supporting the potential of SEVs as promising candidates for α‐Syn detection, analytical quantitation, and subsequent PD diagnosis.

To explore the feasibility of using SEVs for α‐Syn detection and their potential as biomarkers for PD, we employed the differential ultracentrifugation protocol, a method widely recognized as the gold standard for EVs isolation due to its cost‐effectiveness and simplicity [35]. This approach yielded excellent results for SEVs purification, as demonstrated by the characterization profile (Figure 1). Moreover, by using the ELISA technique to analyze α‐Syn_Tot and α‐Syn_Olig we found that SEVs from PD patients carried significantly higher levels of both of them compared to HCs‐derived SEVs (Figure 2). These results align with previous studies that reported increased EVs‐α‐Syn levels in plasma, CSF, and saliva from PD patients compared to HCs [15, 36, 37]. This increase in α‐Syn levels can be explained by the role of EVs in mediating the transfer of α‐Syn aggregates from diseased to healthy neurons, with recipient cells incorporating EV‐bound α‐Syn more efficiently than its soluble form [38, 39]. This mechanism is thought to facilitate the spread of pathological forms of α‐Syn promoting further oligomerization and aggregation of the endogenous protein and thus crucially contributing to neurodegeneration [10, 36, 40, 41]. Additionally, another possible explanation for the increased levels of SEVs‐α‐Syn in PD patients could be attributed to primary lysosomal dysfunction, or to other cellular storage malfunction, as lysosomal inhibition has been shown to significantly enhance the release of EVs‐α‐Syn in cellular models [15, 42, 43, 44].

The results of ROC analysis included in this study indicate a moderate/good diagnostic accuracy of testing SEVs‐α‐Syn for PD with ELISA. Specifically, we report that α‐Syn_Olig load in SEVs shows 65% sensitivity and 75% specificity, with an AUC of 0.71, supporting its role as a molecular biomarker for PD [36, 45, 46]. Taken together, these findings highlight that SEVs‐α‐Syn not only reflect central nervous system pathology more accurately than freely circulating α‐Syn but also provide a valuable diagnostic tool for PD [14, 47]. These comparable AUC values suggest that α‐Syn_Olig in SEVs is a promising biomarker for PD diagnosis, offering a non‐invasive alternative with diagnostic performance comparable to that of CSF and blood‐based assays.

This study revealed a negative correlation between SEVs‐α‐Syn_Olig and cognitive impairment, with lower concentration observed in patients not presenting cognitive deficits (MMSE < 26). To the best of our knowledge, we are among the first to demonstrate that SEVs‐α‐Syn_Olig correlates with MMSE scores, highlighting its potential as an indicator of broader neurodegenerative changes beyond motor symptoms. Interestingly, Sekiya et al. [48] reported elevated α‐Syn_Olig levels in the hippocampus of patients with cognitive impairments, corroborating the validity of our findings and supporting the link between α‐Syn aggregation and cognitive decline. Moreover, our observation about the increase, although not‐significant, of α‐Syn_Olig levels in patients showing higher disease severity (MDS‐UPDRS part III; p = 0.05) is suggestive of the idea that the accumulation of oligomers can contribute to motor symptom progression. Although the correlation between α‐Syn_Olig levels and MDS‐UPDRS part III scores did not reach the established significance threshold, the observed effect size was moderate thus suggestive of a potential relationship that warrants further investigation with a larger cohort size. Similarly, our observations on the slight but not‐significant effects of LEDD on SEV‐α‐Syn are also worthy of future analyses on an expanded cohort of patients.

Notably, our findings are consistent with previous reports linking non‐EV‐associated α‐Syn oligomers with motor symptoms [49]. Altogether, these results support the idea that α‐Syn_Olig accumulation, including that carried by SEVs, can be considered a marker of disease progression and suggests its role in the pathophysiology of PD.

In addition, we did not observe any correlations between α‐Syn_Tot and clinical parameters, nor changes in α‐Syn_Tot levels across PD severity. This lack of association might be attributed to the intricate dynamics of α‐Syn turnover, secretion, solubility, and aggregation in the brain [43]. Given these findings, our follow‐up analysis, focused exclusively on α‐Syn_Olig, demonstrated a stronger link to disease progression and cognitive impairment.

Our is the first longitudinal study assessing SEVs‐α‐Syn_Olig as a prognostic biomarker for PD progression. Intriguingly, we found that both α‐Syn_Olig levels and clinical parameters did not change in patients tested in a 1‐year follow‐up study, underlying the effectiveness of using SEVs α‐Syn_Olig as a sensitive prognostic biomarker. Nevertheless, further investigations are needed by leveraging on larger samples number and more extended follow‐up intervals, as previous studies have shown slight increases in plasma neuronal EVs‐α‐Syn levels in patients with motor progression over 2 years, to conclusively establish SEVs‐α‐Syn_Olig as prognostic PD biomarker [50].

In conclusion, this study strongly supports the potential of exploiting SEVs‐α‐Syn, using a cost‐effective ELISA technique, as a non‐invasive biomarker test for PD diagnosis bearing good sensitivity and specificity. The observed variations in α‐Syn levels across cognitive impairment and PD phenotypes suggest its potential as a reliable prognostic biomarker. These findings establish a solid foundation for further exploration of SEVs in neurodegenerative research and diagnostics, while highlighting the importance of accessible and repeatable tools for both the diagnosis and prognosis of PD and other related disorders.

Author Contributions

A.G., V.C., M.Z. and M.G. conceived and designed the study; M.M.T., C.L., S.B., L.L. and M.Z. recruited and clinically evaluated the patient's cohort; A.G., V.C., and D.C. performed the experiments, G.C. supervised experimental setup; A.G., V.C., M.M.T., M.Z. and M.G. analyzed the data and wrote the manuscript.

Funding

This work was supported by Bando di ricerca traslazionale 2020, Department of Neuroscience “Rita Levi Montalcini”, University of Torino (to M.Z. and M.G.), and a research grant from the University of Pennsylvania Orphan Disease Center (MDBR‐24‐30‐CDKL5) in partnership with the International Foundation of CDKL5 Research (to M.G.).

Consent

The reported cross‐sectional and longitudinal study was approved and supervised by the Ethical Committee of the University of Turin (N° 0040679). Informed consent was obtained from the participants.

Conflicts of Interest

The authors declare no conflicts of interest.

Gurgone A., Cardinale V., Tangari M. M., et al., “Clinical Validation of α‐Synuclein in Salivary Extracellular Vesicles as a Biomarker for Parkinson's Disease: A Longitudinal Study,” European Journal of Neurology 33, no. 2 (2026): e70508, 10.1111/ene.70508.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.