Background

Dementia with Lewy bodies (DLB) and Parkinson’s disease (PD) are Lewy body disorders (LBD) characterized by α-synuclein aggregation that leads to a variety of symptoms, including parkinsonism and dementia secondary to neuronal loss in the cerebral cortex or the brainstem1. Although not entirely elucidated, the pathogenesis of LBD involves multiple mechanisms, including mitochondrial and lysosomal dysfunction, iron dysregulation and inflammatory changes2. REM sleep behavior disorder (RBD) is characterized by dream-enacting behaviors during REM sleep due to the absence of muscle atonia3,4,5. Longitudinal studies have consistently demonstrated that patients diagnosed with the idiopathic form of RBD (IRBD) have a high likelihood of subsequently developing an LBD. The conversion risk to manifest either DLB or PD is 34% within the initial five years after IRBD diagnosis, 74% after ten and 91% after fourteen years6,7,8.

Additionally, post-mortem neuropathological findings in a brain with the clinical diagnosis of IRBD revealed the presence of widespread Lewy pathology, including Lewy bodies and neurites after 10 years of IRBD, but without clinical evidence of either PD or DLB, suggesting that neurodegeneration starts before showing neurological symptoms9. Dopamine transporter (DAT) imaging revealed that almost 60% of IRBD patients had a DAT deficit in the putamen, and this hypometabolism predicts the development of a synucleinopathy within the next three years in 25% of IRBD patients5,10. Also, the analysis of cerebrospinal fluid (CSF) using seeding amplification assay (SAA) in IRBD patients demonstrated that up to 90% of the patients are positive for misfolded α-synuclein in the CSF as detected by RT-QuIC11. Altogether, these findings strongly support that IRBD can be considered a prodromal manifestation of PD or DLB, making these individuals ideal candidates for evaluating biomarkers and future neuroprotective therapeutic strategies.

Blood has emerged as a promising source for examining circulating molecules and cell-based biomarkers. The crucial role of platelets extends beyond their involvement in hemostasis since only 2–6% of them are required for coagulation12. Regarding neurodegenerative diseases, it has been proposed that they might represent a crucial link between blood and brain13 since platelets are a significant source of the amyloid precursor protein (APP) in Alzheimer’s disease (AD), contain α-synuclein, express various neuronal receptors and enzymes, and possess an enzymatic pathway similar to dopaminergic neurons, storing and releasing neurotransmitters14,15,16. Furthermore, platelets contain diverse RNAs, including mRNA and miRNAs. mRNAs are involved in de novo translation and post-transcriptional gene regulation17, and the correlation between the platelet transcriptome and proteome has also been reported18. Due to the presence of a functional miRNA pathway, miRNAs regulate mRNA levels19. Notably, external stimuli can induce modifications in their protein profile, leading to morphological and structural changes that enable platelets to adapt to their environment, underscoring their role as a systemic tool that connects external factors with inner body homeostasis.

A recent study of ours showed that platelet miRNAs could be suitable biomarkers for DLB since we found a 7-miRNA biosignature that distinguished DLB from AD20. Additionally, we observed that, compared with controls, five miRNAs were down-regulated in DLB (hsa-miR- 139 - 5p, hsa-miR- 142 - 3p, hsa-miR- 150 - 5p, hsa-miR- 146a- 5p, hsa-miR- 26b- 5p), six up-regulated in AD (hsa-miR- 132 - 5p, hsa-miR- 146a- 5p, hsa-miR- 25 - 3p, hsa-miR- 142 - 3p, hsa-miR- 26b- 5p, hsa-miR- 6747 - 3p), and two down-regulated in PD (hsa-miR- 128 - 3p, hsa-miR- 139 - 5p)20. Also, other studies have reported that hsa-miR- 136 - 3p and hsa-miR- 409 - 3p might represent risk predictors for PD in CSF21,22.

Since IRBD represents, in most cases, the prodromal stage of LBD and platelet hsa-miR- 142 - 3p, hsa-miR- 146a- 5p, hsa-miR- 150 - 5p, hsa-miR- 26b- 5p are differentially expressed in DLB, hsa-miR- 128 - 3p, hsa-miR- 136 - 3p and hsa-miR- 409 - 3p in PD, and hsa-miR- 139 - 5p in both, in this pilot study we aimed to investigate whether the expression of these miRNAs is also altered in platelets from individuals with IRBD. Additionally, we wanted to know which platelet-related pathways could be affected by miRNA expression changes and identify the corresponding miRNA interactome gene sets. Accordingly, we analyzed the expression of ROCK2, RAC1, WASL and TWF1, genes involved in cytoskeleton stability and dynamics, to identify possible differences between IRBD and controls.

Materials and methods

Participants

Peripheral blood samples collected in Na-citrate tubes were obtained from individuals with IRBD during the first trimester of 2020 at the Sleep Unit of the Hospital Clinic de Barcelona, Barcelona, Spain. Diagnosis of IRBD was based on diagnostic criteria in subjects reporting nightmares and dream-enacting behaviors in whom nocturnal video-polysomnography showed increased electromyographic activity in REM sleep23. Individuals with RBD potentially caused by medication were excluded from the study.

At the time of the diagnosis of IRBD, none of the patients reported motor and cognitive problems, and the neurological exam revealed no signs of parkinsonism or cognitive impairment. Demographic and clinical data were obtained during the initial visit and follow-up visits, which took place every 3–12 months. Once neurological signs or symptoms were noted in any of the follow-up visits, patients were classified as PD or DLB according to the diagnostic criteria24,25.

Patients were requested to provide blood samples for research during their regular appointments. For this study, patients were classified by their neurological condition as either disease-free IRBD (IRBD-IRBD) or IRBD that phenoconverted to PD (IRBD-PD) or DLB (IRBD-DLB). The onset of RBD was determined based on the date their spouses started to notice abnormal sleep behavior. The duration of IRBD was calculated as the interval between estimated RBD onset and blood sample collection.

Control blood samples (CTRLs) were obtained from healthy individuals without a family history of neurological disorders and without complaining of intense dream-related behaviors suggestive of IRBD or any symptoms or signs indicating parkinsonism or cognitive impairment. They were recruited from the Department of Neurology at Hospital Germans Trias i Pujol (Badalona, Barcelona, Spain).

The study received ethical approval from the local Ethical Committees for Clinical Investigation at the participating institutions, and all participants provided written informed consent.

Purification of platelet-derived RNA and reverse transcription

Blood samples were obtained in Na-citrate Vacutainer tubes (BD, Plymouth, UK), and to avoid expression changes due to platelet activation, samples were processed within the first four hours after blood extraction. First, platelet-rich-plasma was obtained by centrifugation at 200 g for 15 min, and platelet-rich pellets (PRP) after a second centrifugation at 2.500 g for 15 min. PRP was stored at − 80 °C until RNA purification and thawed on ice before processing. Total RNA and miRNA isolation were performed using the mirVana™ miRNA Isolation Kit (ThermoFisher, Waltham, MA, USA), and isolated miRNAs and RNAs were stored at − 80 °C until expression analysis.

Reverse transcription of 150 ng of total platelet RNA was performed with the Ready-to-go™ You-Prime First-Strand Beads (GE Healthcare, Buckinghamshire, UK), adjusted to a final volume of 32 µl with DEPC-treated H2O. Samples were heated at 65 °C for 10 min and chilled on ice for 2 min. RNA was transferred to the tube with first-strand reaction mix beads, and 1 µl of random primers (10ng/ul) was added. After incubating for one minute at room temperature, mixes were gently vortexed and incubated at 37 °C for 60 min.

Reverse transcription of small RNAs was carried out using the MiRCURY LNA RT kit (Qiagen Hilden, Germany), where 2 µl containing 10 ng of each sample was mixed with 0.5 µl of UniSp6 and 4.5 µl of H2O, 2 µl of 5x buffer and 1 µl of reverse transcriptase included in the kit. The mixes were incubated for 60 min at 42 °C, the reverse transcriptase was inactivated for 5 min at 95 °C, and samples were cooled to 4 °C immediately.

MiRNA quantification

Quantitative PCR (qPCR) was performed on a LightCycler 480 (Roche, Basel, Switzerland). miRCURY LNA miRNA Custom PCR Panels (Qiagen) with synthetic UniSp3 as interplate calibrator and UniSp6 as retrotranscription control and for the normalization of the results were used to analyze the expression of the following miRNAs: hsa-miR- 139 - 5p (deregulated in both DLB and PD)20, hsa-miR- 142 - 3p, hsa-miR- 146a- 5p, hsa-miR- 150 - 5p, hsa-miR- 26b- 5p (DLB-specific miRNAs)20, hsa-miR- 128 - 3p20, hsa-miR- 136 - 3p and hsa-miR- 409 - 3p (PD-specific)21,22. Hsa-miR- 191 - 5p was included as a potential housekeeping miRNA, as suggested by the provider.

qPCRs were performed after the dilution of 2 µl of cDNA in a total volume of 200 µl of which 75 µl were combined with 75 µl of 2x PCR Master mix included in the kit. The final volume applied to each well was 10 µl. After sealing, plates were centrifuged at 1500xg for 1 min, and real-time PCR was performed after polymerase activation at 95 °C for 2 min, during 45 amplification cycles of 95 °C for 10 s and 56 °C for 1 min. Fluorescence was acquired at the end of each extension step, and the final denaturation was from 55º − 95º, with a ramp of 4.4 s.

MiRNA target prediction and analysis

Possible mRNA targets of deregulated miRNAs were predicted using miRTarbase26, accepting as target genes those reported by studies conferring strong evidence. Each miRNA interactome was analyzed by functional enrichment analysis and building protein-protein interaction (PPI) networks using the String DataBase27. Gene products were checked for their presence in platelets at the PlateletWeb - Systems Biology Workbench28.

Establishment of housekeeping genes in platelets

To establish housekeeping genes (HKG) in platelets, the expression levels of seven candidate mRNAs (SNRPD, RRAS, RDX, PP1 A, B2M, GADPH, DDX23) were tested in AD (n = 8) and LBD patients (n = 10), and controls (n= 12) (all individuals previously described20). Quantification was performed by qPCR, as described in the next section and using the primers specified in Table 1. ΔCt values were calculated by pair-wise comparison of all genes. The mean ΔCt value of the control group was used to calculate the ΔΔCt value by the ΔΔCt-method29,30. The selection of HKG was based on the following criteria: (1) Ct values from individuals within each group followed a normal distribution, (2) no significantly different expression in any of the groups (expression change between 0.75 and 1.25), and (3) minimal differences in Ct values between the groups (close between − 0.5 and 0.5). The normality of the distribution of the Ct values for each gene was tested using the Shapiro test. Furthermore, the suitability of the selected HKGs was also checked using the NormFinder method in R software (ReadqPCR and NormqPCR packages)31.

Table 1 cDNA primer sequences for HKG candidates.
Full size table

mRNA quantification

Relative mRNA expression of the genes identified by target analysis and the corresponding HKGs was assessed using real-time PCR with the Luna Universal qPCR Master Mix kit (New England BioLabs, Ipswich, MA, USA) and carried out on a Rotor-Gene 6000 (Qiagen). PCR was performed in 15µL reactions, containing 7.5µL of Luna Universal qPCR Master Mix, 10µM of each primer (Table 2) and 1µL of cDNA. The cycling program consisted of an initial 1-minute denaturation step followed by 40 cycles of 10 s at 95 °C, 20 s at the primer-specific annealing temperature, and 30 s at 60 °C. Annealing temperature of 57 °C was used for the amplification of RRAS and RDX, 58 °C for SNRPD, B2M, GADPH, DDX23, RAC1 and ROCK2 and of 60 °C for TWF1, WASL and PP1 A. Fluorescence was acquired at the end of each extension step, and the final denaturation was from 55º − 95º, with a ramp of 4.4 s. All samples were analyzed in duplicate to ensure accuracy, and the assays were performed independently to confirm reproducibility.

Table 2 cDNA primer sequences for RAC1, ROCK2, TWF1 and WASL.
Full size table

ΔΔCt method

Analyses of miRNA and mRNA expression were performed separately, and Ct values were averaged between duplicates. The normalization factor for miRNAs was derived from the geometric mean of the UniSp6 Ct values32,33. For mRNAs, the normalization factor was derived from the geometric mean of Ct values of RDX and SNRPD established as HKGs in platelets. ΔCt was obtained for all samples, the mean of ΔCt values from the control group was used to calculate the ΔΔCt value for each sample, and the expression change was calculated by 2−ΔΔCt.

Statistical analysis

Expression data were corrected by age and sex using a multiple linear regression method. Possible expression changes for IRBD vs. controls and clinical parameters were analyzed using the Wilcoxon-Mann-Whitney test. The correlation between miRNA expression levels and clinical variables (sex, age and estimated RBD duration) was determined by multiple and linear model analysis. The normal distribution of residuals was verified using a Q-Q plot. Since data followed a normal distribution, Pearson correlation coefficient R and its corresponding p-value were calculated. Post-hoc power analysis, where values > 0.8 are accepted as optimal, was performed for those miRNAs and mRNAs that showed significant expression changes. Statistical analyses and plots were carried out using R software and stats, ggpubr, ggplot2 and pwr packages, version 4.2.3 for Windows.

Results

Demographic and clinical data

Demographic and clinical data of patients are shown in Table 3. The mean age of IRBD patients and CTRLs was similar. There were more males in the IRBD group than in CTRLs. At the time of blood sampling, 20 out of the 29 (69.0%) IRBD patients remained free of any neurological disorder (IRBD-IRBD), while nine (31.0%) had phenoconverted to PD (IRBD-PD, n = 5) or DLB (IRBD-DLB, n = 4).

Table 3 Demographic and clinical data of the participants of the study.
Full size table

MiRNA expression in platelets

The expression of nine miRNAs (hsa-miR- 139 - 5p, hsa-miR- 142 - 3p, hsa-miR- 146a- 5p, hsa-miR- 150 - 5p, hsa-miR- 26b- 5p, hsa-miR- 128 - 3p, hsa-miR- 136 - 3p, hsa-miR- 409 - 3p, hsa-miR- 191 - 5p) was analyzed in 29 IRBD patients and 20 CTRLs. The mean Ct and standard deviation values before correction are shown in Table 4. Of these, hsa-miR- 139 - 5p and hsa-miR- 142 - 3p were significantly diminished in the 29 IRBD platelet samples compared to CTRLs (p = 0.010 and p = 0.017, respectively), with power values of 0.861 and 0.981 after post-hoc power analysis, respectively (Fig. 1). Hsa-miR- 191 - 5p expression was significantly increased in IRBD compared to CTRLs (p = 0.023; power value = 0.813) (Fig. 1). The expression of the other six miRNAs did not significantly differ between IRBD and CTRLs.

Table 4 Mean Ct values and standard deviations of all MiRNAs.
Full size table
Fig. 1
Fig. 1The alternative text for this image may have been generated using AI.
Full size image

Comparison of hsa-miR- 139 - 5p, hsa-miR- 142 - 3p and hsa-miR- 191 - 5p expression levels in platelets between IRBD and controls. P-value obtained by Wilcoxon-Mann-Whitney test.

When the IRBD group was divided into IRBD-IRBD, IRBD-PD and IRBD-DLB, the diminution of hsa-miR- 139 - 5p and hsa-miR- 142 - 3p and increase of hsa-miR- 191 - 5p remained significant compared to CTRLs (Fig. 2) in IRBD-IRBD. The expression did not differ significantly in IRBD-PD and IRBD-DLB, compared to IRBD-IRBD or controls (Fig. 2).

Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

Comparison of hsa-miR- 139 - 5p, hsa-miR- 142 - 3p and hsa-miR- 191 - 5p expression levels in platelets between IRBD-IRBD, IRBD-PD and IRBD-DLB subgroups and controls. P-value obtained by Wilcoxon-Mann-Whitney test.

Correlation between MiRNA expression and clinical variables

Then, we analyzed the possible correlation between the expression levels of the three significantly deregulated miRNAs and IRBD duration before phenoconversion (for IRBD-PD and IRBD-DLB) or duration (for IRBD-IRBD) and age at RBD onset. No correlation between platelet miRNA expression levels and the age of RBD onset (Pearson correlation coefficient R between − 0.16 and − 0.02) or RBD duration (Pearson correlation coefficient R between − 0.29 and − 0.20) was found after analyzing all 29 IRBD subjects. However, when correlation analysis was carried out considering three groups, IRBD-IRBD, IRBD-PD and IRBD-DLB, a strong negative correlation was found in IRBD-PD between miRNA expression levels and both age at RBD onset (Fig. 3A) and RBD duration before phenoconversion (Fig. 3B).

Thus, miRNA expression was higher in cases that developed PD before the age of approximately 62 years and in cases that converted to PD within the first 8 years of RBD (Fig. 3B). Due to the small sample size, the p-value for Pearson correlation could not be assessed.

MiRNA expression did not differ significantly between genders.

Fig. 3
Fig. 3The alternative text for this image may have been generated using AI.
Full size image

Correlation between (A) miRNA expression and age at RBD onset, and (B) RBD duration before phenoconversion. Pearson correlation coefficient (R) is shown in each graph.

Analysis of MiRNA interactomes

Then, the interactome gene sets of hsa-miR- 142 - 3p, hsa-miR- 139 - 5p and hsa-miR- 191 - 5p were analyzed. The hsa-miR- 142 - 3p interactome comprised 39 genes with a significant PPI enrichment (p = < 0.0001). The gene products were searched for their presence in platelets, and 15 had been reported to be expressed either at the mRNA or the protein level25. These were analyzed by String showing a significant PPI enrichment (p = 0.0105) (Fig. 4A). Nine genes were predicted to be components of the cytoskeleton (GO: 0005856, FRD: 0.0069), three of these are involved in actin polymerization or depolymerization (GO: 0008154, FRD: 0.0158), and three in the G13 signaling pathway (WP524, FRD: 0.0027) (Fig. 4B).

Fig. 4
Fig. 4The alternative text for this image may have been generated using AI.
Full size image

The hsa-miR- 142 - 3p interactome (A) of the proteins expressed in platelets. (B) Proteins located in the cytoskeleton (blue), and involved in actin polymerization/depolarization (red) or G13 signaling (green).

The hsa-miR- 139 - 5p interactome gene set also contained 39 genes. Of these, 19 were expressed in platelets, and String analysis showed a highly significant PPI enrichment (p = 1.08e- 08). However, no biological process or molecular function related to platelet function was found, and KEGG Pathways were associated with several cancer types.

The hsa-miR- 191 - 5p interactome gene set contained 19 genes. Eight were expressed in platelets, but no PPI enrichment was found (p = 0.165), and KEGG pathways were also related to cancer.

Establishment of housekeeping genes in platelets

Of the seven genes initially tested, SNRPD and RDX fulfilled the criteria for HKG definition: (1) Shapiro test p-values were not significant (0.1039 and 0.3385, respectively), meaning that Ct values followed a normal distribution (Fig. 5); (2) mean and range of expression change comparing CTRLs with LBD and AD, and LBD with AD assessing both genes were between 0.75 and 1.25, indicating that there were no significant differences between CTRL, AD and LBD groups; (3) the differences between Ct values were close to 0 (SNRPD-RDX: − 0.18 in CTRLs vs. AD, − 0.18 in CTRLs vs. LBD, and 0 in AD vs. LBD). Furthermore, SNRPD and RDX had the lowest rho values after performing NormFinder method (0.186 and 0.189, respectively).

Fig. 5
Fig. 5The alternative text for this image may have been generated using AI.
Full size image

Distribution of Ct values of SNRPD and RDX gene expression for all samples. Each bar represents the number of patients corresponding to each Ct value. Samples of all three groups were included (n = 30). Normal distribution was tested with the Shapiro test.

Expression of hsa-miR- 142 - 3p regulated genes in platelets

Based on the hsa-miR- 142 - 3p interactome analysis, we further analyzed the four genes which were components of the cytoskeleton and at least one additional pathway: ROCK2, RAC1, WASL and TWF1 (Fig. 4) in all IRBD patients and 31 CTRLs. Whereas platelet ROCK2, RAC1 and TWF1 expression did not differ between IRBD and controls, WASL expression was significantly increased in IRBD (1.58 ± 0.70 vs. 1.08 ± 0.40 in CTRLs, p = 0.008 and power value = 0.957; Fig. 6A). Additionally, a significant negative correlation was found between hsa-miR- 142 - 3p and WASL expression, where patients with lower platelet hsa-miR- 142 - 3p levels showed higher WASL expression (p = 0.001; Fig. 6B).

Fig. 6
Fig. 6The alternative text for this image may have been generated using AI.
Full size image

WASL expression in platelets. (A) Comparison of WASL expression levels between IRBD and controls; p-value obtained by Wilcoxon-Mann-Whitney test (B) Correlation between WASL and hsa-miR- 142 - 3p expression. Pearson correlation coefficient and p-value obtained by linear regression. Pearson correlation coefficient (R) and the corresponding p-value are shown.

Discussion

In the present study, we analyzed the expression of nine miRNAs in platelets of patients diagnosed with IRBD. Six of these miRNAs were selected from a previous study, where we found that hsa-miR- 139 - 5p was diminished in both PD and DLB, hsa-miR- 146a- 5p, hsa-miR- 150 - 5p, hsa-miR- 26b- 5p, and hsa-miR- 142 - 3p were diminished in DLB whereas hsa-miR- 128 - 3p was diminished in PD compared with healthy controls20. Considering the previous findings, we aimed to determine whether these miRNAs are already deregulated in IRBD, a prodromal condition for LBD7 and whether they could be useful predictors of IRBD progression to DLB or PD. We also analyzed two additional miRNAs, hsa-miR- 136 - 3p and hsa-miR- 409 - 3p, which were described as differentially expressed in PD compared with controls21,22.

As hsa-miR- 191 - 5p had been used as a housekeeping miRNA in whole blood, it was also assessed in our study, but its suitability as a housekeeping gene in platelets could not be confirmed. Instead, we found that hsa-miR- 191 - 5p was overexpressed in IRBD compared to CTRLs. Additionally, two of the previously identified miRNAs, hsa-miR- 142 - 3p and hsa-miR- 139 - 5p, were decreased.

The pooling of the data from our work on platelet-derived miRNAs in synucleinopathies, including a previous study20 and the current one, seems to reveal the following pattern: in IRBD, hsa-miR- 139 - 5p and hsa-miR- 142 - 3p are diminished, while hsa-miR- 191 - 5p is increased. Once phenoconverted to DLB, the disease-specific miRNA signature in platelets includes hsa-miR- 139 - 5p, hsa-miR- 142 - 3p, hsa-miR- 146a - 5p, hsa-miR - 150 - 5p and hsa-miR- 26b - 5p. In contrast, in those IRBD patients who phenoconverted to PD within less than eight years of IRBD, hsa-miR- 191 - 5p was markedly increased, and upon phenoconversion to PD, hsa-miR- 139 - 5p and hsa-miR- 128 - 3p are diminished. These defined profiles suggest that deregulation of hsa-miR- 142 - 3p, hsa-miR- 139 - 5p and hsa-miR- 191 - 5p is characteristic of IRBD, with hsa-miR- 191 - 5p overexpression potentially indicating an elevated risk of progression to PD.

Role of hsa-miR- 142 - 3p and WASL in platelet morphology and function

Hsa-miR- 142 - 3p plays a crucial role in megakaryocyte maturation and the regulation of platelet biosynthesis, including proplatelet formation. This miRNA is abundant in platelets, regulating the expression of actin-modulating genes involved in cytoskeletal dynamics34. In vivo studies in mice lacking this miRNA have shown reduced platelet production and impaired maturation, although the amounts of megakaryocytes were increased34.

Platelet activation is accompanied by the rapid expansion of the platelet cytoskeleton due to the increase in actin polymerization mediated by G protein-coupled receptors (GPCRs)35, and our interactome analysis revealed that hsa-miR- 142 - 3p regulates 15 platelet mRNAs, nine of which encode cytoskeleton proteins. Three, RAC1, WASL and TWF1, are involved in actin polymerization/depolymerization, and three (ROCK2, RAC1 and WASL) are involved in G13 signaling. Notably, we observed WASLoverexpression in IRBD, and its negative correlation with hsa-miR- 142 - 3p expression supports a regulatory relationship, as reported for megakaryocytes35 and human monocytes36. WASL encodes an actin nucleation factor that regulates actin polymerization via interaction with the ARP2/3 complex, which is crucial for proplatelet formation37. The overexpression of WASL we observed in platelets could lead to an increased platelet volume without initiating platelet activation. Accordingly, changes in platelet morphology, including increased volume, have been reported in PD platelets38,39.

Furthermore, WASL has been associated with long actin-microspike formation, neurite extension, dendrite spine morphogenesis in hippocampal neurons40 and the regulation of oligodendrocyte myelination41. Genetic variants of WASL were described in a family with early onset PD and were associated with an altered neurite branching pattern, lower ROS tolerance and higher alpha-synuclein levels in neurons42. These results suggest that WASL dysfunction in the brain might contribute to synucleinopathies, and our findings indicate that hsa-miR- 142 - 3p downregulation in IRBD could lead to altered platelet morphology, potentially representing an early stage of synucleinopathies linked to systemic inflammation43.

Potential role of hsa-miR- 139 - 5p in neuroinflammation

Whereas reduced hsa-miR- 139 - 5p was reported in the blood of PD patients44, the levels of free circulating hsa-miR- 139 - 5p were increased in the plasma of idiopathic PD45. This miRNA promotes neuronal survival and regulates synaptic plasticity and nerve regeneration in the hippocampus46. It also regulates PAFAH1B, a gene that codifies for the non-catalytic subunit of platelet-activating factor acetylhydrolase isoform IB47 and inactivates the platelet-activating factor (PAF). Low PAF levels, caused by hsa-miR- 139 - 5p deficiency and subsequent PAFAH1B overexpression, can trigger inflammatory cascades48, suggesting a link between miRNA deregulation, platelet dysfunction and neuroinflammation49. Our current findings suggest that an altered inflammatory response could be an early event during the development of LBD. In fact, a study in IRBD using neuroimaging PET tracers has shown increased microglial activation in the substantia nigra, suggesting that inflammation may be early involved in the neurodegenerative LBD process50.

Biomarker potential of hsa-miR- 191 - 5p

In addition to deregulating the above-discussed miRNAs, we found elevated hsa-miR- 191 - 5p expression in platelets of IRBD patients, especially in those who converted to PD a few years after RBD onset. Although initially proposed as a housekeeping miRNA, hsa-miR- 191 - 5p may have functional roles in disease development. This miRNA is abundantly expressed in the brain51 and enriched in vesicles derived from platelets and erythrocytes of healthy individuals52. Increased hsa-miR- 191 - 5p levels have also been observed in the plasma of other neurodegenerative diseases, such as AD and multiple sclerosis53,54. Since this miRNA regulates T- and B-cell mediated immune response55,56, its upregulation in IRBD may contribute to neuroinflammation.

Our data suggest that IRBD patients with more than a 2-fold increase in hsa-miR- 191 - 5p expression could have a higher risk of phenoconversion to PD within the first eight years of IRBD progression. However, due to the limited number of participants and phenoconvers to PD, further research is necessary to validate the role of hsa-miR- 191 - 5p as a predictive LBD biomarker.

Study limitations and future directions

Our study has two main limitations. The first is the sample size, which limited our ability to subdivide the IRBD cohort into groups based on phenoconversion. Correspondingly, there was no sufficient power for the complete statistical analysis of IRBD-PD and IRBD-DLB. However, the results indicate that has-miR- 191 - 5p should be further studied in IRBD. The second limitation was the lack of an adequate endogenous miRNA control in platelets. Although we have identified RDX and SNPRD as suitable housekeeping genes to control mRNA expression, our effort to identify an appropriate miRNA is still ongoing. Therefore, and as reported in previous studies, we used the exogenous spike-in UniSp6 to normalize the results.

Additionally, similar to PD and DLB, IRBD is more prevalent in men than in women. Correspondingly, 56% of the control individuals but only 17% of the IRBD cohort were female. Larger independent cohorts are necessary to further assess the role of platelet miRNAs as biomarkers for LBD progression. Additional studies in different laboratories are also required to validate our findings.

Conclusion

Of the eight miRNAs diminished in platelets of LBD patients, hsa-miR- 142 - 3p and hsa-miR- 139 - 5p are also decreased in IRBD, suggesting a role in disease progression. Moreover, WASL expression regulated by hsa-miR- 142 - 3p is increased in IRBD, supporting the hypothesis that platelet function is impaired early in prodromal synucleinopathies. If hsa-miR- 191 - 5p overexpression found in IRBD patients could be a predictive biomarker for phenoconversion, needs to be further studied.