2. Neuronal Synucleinopathies
2.1. What We Know from PD Cellular Models
Several in vitro studies suggest a potential role of exosomes as mediators of neurodegeneration in PD. Initial studies in αSyn-expressing SH-SY5Y cells proposed that αSyn is partly released via exosomes in a calcium-dependent manner
[6] and that such exosomes may mediate αSyn transmission between neuronal cells in a manner dependent upon lysosomal activity
[68,69,70][16][17][18]. A subsequent study in primary neurons and neuronal-like cells suggested that exosome-associated αSyn oligomers demonstrate a higher capacity of being taken-up by surrounding cells, as well as, of exerting more toxic effects on recipient cells, compared to exosome-free forms of the protein, uncovering thus a key mechanism in the spread of αSyn pathology in the brain
[7]. In accordance with previous observations, it has been reported that exosomes derived from the mouse neural crest-derived Neuro 2A cell line treated with low concentrations of αSyn pre-formed fibrils (PFFs) provide an environment that favors αSyn oligomerization, as assessed by a continuous Thioflavin T fluorescence assay
[71][19]. To further support this notion, the
resea
uthorchers analyzed the molecular composition of these exosomes and found that certain exosomal lipids, namely ganglioside lipids GM1 and GM2, are sufficient to cause the accelerated aggregation of αSyn.
Other studies have shown that not only neuronal- but also glial-derived exosomes may play a significant role in the progression of PD pathogenesis by delivering potential harmful cargoes to neighboring neuronal or glial cells. Various in vitro studies have proposed that microglia, the primary resident immune cells of the brain, are capable of internalizing and subsequently degrading different conformations of αSyn, pinpointing a leading role of microglia on the onset and progression of PD
[72,73,74,75,76,77][20][21][22][23][24][25]. On the other hand, pathological αSyn can also activate microglia, leading to the secretion of pro-inflammatory cytokines and neuroinflammation events that may underlie PD pathogenesis
[78][26]. Moreover, microglia themselves can also secrete exosomes
[79][27] and the role of microglial-derived exosomes in cell-to-cell transmission of αSyn pathology has recently gained considerable attention. In particular, treatment of BV-2 microglia cells with aggregated αSyn has been shown to increase the release of exosomes, which may further exert neurotoxic effects due to the expression of the MHC class II molecules and membrane TNF-a
[80][28]. Towards the same direction, inoculation of mouse primary microglia cells with human αSyn PFFs has been reported to evoke the release of exosomes containing pro-inflammatory cytokines that further lead to increased αSyn aggregation and pathology transmission in recipient neuronal cells both in vitro and in vivo
[13].
Even more, environmental neurotoxic agents such as manganese (Mn
2+) have been reported to increase neuronal exosome secretion and promote PD progression
[81][29]. Supportive of this notion are data showing that exposure of MN9D dopaminergic cells to Mn results in increased expression of the Rab27a protein that controls the exosome release. Mn is also suggested to impact extracellular exosome-associated miRNAs that are significantly implicated in pathways found to be dysfunctional in PD
[82][30]. Similarly, exosomes derived from methamphetamine-treated SH-SY5Y cells carry pathological αSyn and exosome-mediated transmission of phosphorylated αSyn at Ser129 from neurons to astrocytes is increased upon methamphetamine exposure
[83][31]. Impairment of the autophagy-lysosome pathway, which is closely linked to PD development, has been reported also to influence exosomal release levels. Different studies have shown that pharmacological inhibition of autophagy promotes exosome release and αSyn transmission
[7,84,85][7][32][33]. More specifically, autophagic dysfunction facilitates the release of αSyn via exosomes and the propagation of αSyn-related pathology, which eventually evokes apoptotic cell death of recipient neurons and disease progression
[86][34]. On the contrary, enhancement of autophagy decreases or even inhibits exosome release in cellular models
[87][35].
Exosomes, as mentioned above, are also able to shuttle miRNAs and contribute to PD progression by modulating the expression of target genes in recipient cells
[88,89][36][37]. For example, treatment of primary rat cortical astrocytes with the pro-inflammatory cytokines IL-1β and TNF-a induced the secretion of exosome-associated miR-125a-5p and miR-16-5p that down-regulate the expression of the neurotrophin receptor NTRK3 (TRKC), and its downstream effector protein, Bcl2. Down-regulation of these neuronal targets reduces dendritic growth and complexity, spike rates and burst activity
[63][38]. Moreover, application of secreted astrocytic exosomes on dopaminergic cells increased the protein levels of microvesicle-related miR34a that enhances the susceptibility of SH-SY5Y dopaminergic cells to the PD-linked neurotoxins MPP+ and 6-OHDA through repression of Bcl2 protein
[90][39]. Another study reported that miR-137, which regulates oxidation resistance 1 gene, is upregulated and contributes to the induction of oxidative stress in primary neurons derived from PD mouse models
[91][40]. Moreover, astrocyte-derived exosomal miR-200a-3p has been shown to exert a neuroprotective role through down-regulation of the mitogen-activated protein kinase 4 (MKK4) in SH-SY5Y cells and primary dopaminergic neuron cultures
[14].
Finally, PD-linked genes have also been associated with the regulation of exosome biogenesis and release. Autosomal dominant mutations in
LRRK2 (Leucine-rich repeat kinase 2) gene are associated with both familial and sporadic PD
[92,93][41][42]. In particular, LRRK2 is a kinase strongly implicated in the endolysosomal pathway, thus regulating exosome biogenesis via its interaction with Rab5b and Rab7
[94,95][43][44]. Notably, alterations of LRRK2 expression in primary neurons significantly impaired synaptic vesicle endocytosis, a defect that could be rescued by the simultaneous overexpression of Rab5b
[94][43]. Moreover, phosphorylated LRRK2 at Ser-1292 has been found associated with exosomes isolated from urine of PD patients, suggesting that the LRRK2-dependent formation of exosomes may be linked to sporadic PD
[96,97][45][46]. Another gene related to the development of PD is ATP13A2 (PARK9), which encodes for a transmembrane endolysosomal ATPase, expressed in the pyramidal neurons within the cerebral cortex and the dopaminergic neurons of the substantia nigra
[98][47]. ATP13A2 was detected in MVBs, and its presence seems to be related to the number of ILVs and the release of exosomes in general and of exosome-associated αSyn
[10,99,100][10][48][49]. Similar effects have been reported for the vacuolar protein sorting 35 (VPS35), a component of the retromer complex involved in the recycling of membrane proteins from endosomes to the trans-Golgi network. The D620N mutation in VPS35 gene has been linked to late-onset PD and has been found to cause endosomal alterations and trafficking defects
[101][50]. In addition, mutations in the E3 ligase parkin, which is physiologically implicated in the regulation of mitophagy through the ubiquitination of mitochondrial membrane proteins, alter the organization of MVBs increasing thus the exosome release
[102][51].
2.2. Knowledge Acquired from PD Animal Models
Aging is another factor that may play an important role in exosome-associated αSyn transmission, given that microglia derived from elderly mice have shown to exert deficits in the uptake of exosome-associate αSyn, as compared to microglia derived from young animals
[103][52]. Given that aging is the primary risk factor for developing PD, such data insinuate that a larger number of pathology-related exosomes are “trapped” in the intercellular space of aged mice and this impaired clearance of pathological αSyn species may potentially exert harmful effects on neurons and eventually, may contribute to neuroinflammation and disease progression. Supporting further this hypothesis, a subsequent study showed that the uptake of pathogenic αSyn oligomers bound on exosomes triggered the pro-inflammatory activation of microglial cells and led to the secretion of various cytokines and inhibition of the autophagic machinery, eventually resulting in αSyn accumulation
[12].
Interestingly, misfolded and oligomeric αSyn, or even the PD-linked A53T αSyn mutant seem to be rather associated with exosomal vesicles, leading to the hypothesis that pathological αSyn species can be spread via their interaction with membrane vesicles
[85,104,105][33][53][54]. Moreover, it has been shown that delivery of extracellular vesicles isolated from αSyn-overexpressing HEK293 cells into the striatum of wild-type mice is responsible for the spread of human αSyn throughout the brain
[106][55]. On the contrary, others supported that when exosomes pre-incubated with αSyn pre-formed fibrils (PFFs) were delivered to the striatum of wild-type mice, they neutralized the toxic effects of PFFs, thus raising questions regarding the precise role of exosomes in the development and spread of neuropathological processes
[107][56]. Another study suggested that the release of exosome-associated αSyn is regulated by sumoylation
[108][57]. Recent evidence indicates that exosomes are partially responsible for the transmission of αSyn-related pathology from neurons to astrocytes and that the internalization of αSyn by astroglia induces inflammatory responses
[83][31].
Strikingly, treatment of Prnp-αSyn A53T transgenic mice with conduritol-B epoxide (CBE), a pharmacological inhibitor of the lysosomal enzyme β-glucocerebrosidase (GCase), resulted in elevated levels of exosome-associated oligomeric αSyn species
[109][58]. Additionally, GCase deregulation resulted in high levels of extracellular vesicles in the hemolymph of a Drosophila model of PD
[110][59].
2.3. Lessons Obtained from Patient-Derived Material
The delivery of exosomes isolated from the brain, CSF or blood (plasma or serum) of patients with alpha-synucleinopathies in various cellular and animal models has shed light on their potential role in the aggregation and seeding of aSyn pathology. Analysis of CSF-derived exosomes from PD patients and controls revealed that only PD and not control exosomes evoke αSyn oligomerization when applied to human H4 neuroglioma cells, probably due to the presence of aberrant protein species that are distinct from those contained in exosomes derived from control subjects
[111][60]. Notably, microglial-derived exosomes have been identified in the CSF of PD and MSA patients, triggering αSyn aggregation in vitro
[13]. Additionally, it has been shown that plasma exosomes derived from PD patients can be taken up and subsequent activate microglia in vitro. Patient-derived plasma exosomes were found to inhibit autophagy in BV2 cells and subsequently accelerate aggregation and secretion of αSyn
[12].
Interestingly, it has been shown that microglial cells were more prone to internalize exosomes in vivo, rather than neuronal or astroglial cells, in an experimental set-up where exosomes derived from PD patients’ plasma were injected in the striatum of 8-month-old mice
[12]. Furthermore, it has been reported that neuron-derived exosomes isolated from the brain, CSF or blood of PD or DLB patients can trigger αSyn aggregation in both in vitro and in vivo models
[69,111,112,113][17][60][61][62]. Even more, likely neurogenic L1CAM-purified extracellular vesicles (EVs) derived from the plasma of PD patients have been recently shown to increase the accumulation of αSyn in the midbrain and to accelerate the progression of PD pathology in the Prnp-αSyn A53T transgenic mice
[114][63]. EV-associated miRNA profiling revealed that the novel miRNA, novel_miR_44438, was expressed at significantly higher levels in neurogenic EVs from PD subjects than in healthy controls, inhibiting neuronal exosome release and αSyn efflux through the NDST1-HS pathway
[114][63].
Since exosomes are considered as potential carriers of toxic proteins, their role in PD pathogenesis has been extensively studied via the measurement of their levels in patient-derived biological fluids. Specifically, it has been demonstrated that although the levels of αSyn in the CSF of PD patients were lower compared to controls, the exosome-associated αSyn in the plasma of these patients was significantly higher
[115,116,117,118][64][65][66][67]. As mentioned above, PD pathogenesis is closely linked to lysosomal function, with up to 7% of patients carrying a loss-of-function mutation in the GBA1 gene that encodes for the lysosomal enzyme β-glucocerebrosidase (GCase)
[119,120,121][68][69][70]. Finally, it is worth-mentioning that the ratio of exosome-bound αSyn/total αSyn in the plasma of patients with PD is related to GCase enzymatic activity
[122][71].
2.4. Extracellular Vesicles in DLB
In contrast to the plethora of studies investigating the role of extracellular vesicles (or exosomes in particular) in both the pathogenesis and the prognosis of PD, there is a paucity of data regarding the implication of these nanovesicles in DLB. Only a few studies have focused on the involvement of extracellular vesicles in the pathogenesis of the disease or their role as potential biomarkers for DLB. Specifically, it has been originally reported that CSF-derived exosomes from DLB patients are enriched in αSyn, which can be spread and trigger αSyn oligomerization in a dose-dependent manner
[111][60]. These findings were further supported when one year later it was reported that brain exosomes isolated from DLB patients containing Aβ, tau and αSyn, led to the recruitment of tau and αSyn into the formation of phosphorylated protein aggregates in non-diseased rodent brains
[112][61]. Such exosomes seem to be internalized preferably by neurons and secondarily by astrocytes via a Rab5-mediated endocytosis, thus participating in the establishment of αSyn pathology
[112][61]. Similarly, both PD- and DLB-derived exosomes were found to induce the oligomerization of soluble αSyn in recipient cells
[7]. Interestingly, the number of CSF-derived exosomes and the exosome-associated αSyn cargo isolated from DLB patients were found significantly lower when compared to PD patients
[111][60].
DLB and AD are the most common types of dementia and the identification of valid biomarkers that could distinguish between these two diseases is an unmet need. To this end, it has been proposed that the plasma microRNA expression profile could be utilized for the differential diagnosis of the two diseases, given that hsa-miR-451a and hsa-miR-21-5p were significantly down regulated in AD as compared to DLB cases
[123][72]. Likewise, a more recent study demonstrated a significantly reduced expression of various pro-inflammatory genes and the down-regulation of several inflammatory pathways in DLB serum EVs, thus proposing that these vesicles could serve as potential diagnostic markers for DLB
[124][73]. Moreover, in the same study the down-regulation of the protein ubiquitination pathway-related genes UBE3A, USP47, and PSMD4, or even of the ATP-binding cassette family genes ABCA7 and ABCA13 in DLB-derived EVs, further suggested their use as potential diagnostic biomarkers for DLB
[124][73]. However, additional studies are required to identify blood-based diagnostic biomarkers for DLB and related synucleinopathies, since various methodological challenges need to be currently addressed.
3. Exosomes in MSA
MSA, albeit less prevalent, pertains to the broad spectrum of alpha-Synucleinopathies
[125][74]. As distinct from PD and DLB, it is widely acknowledged that in MSA the aggregation of pathological αSyn species takes place primarily in the cytoplasm of oligodendrocytes and secondarily in neurons, thus leading to the formation of glial cytoplasmic inclusions (GCIs) and neuronal cytoplasmic inclusions (NCIs), respectively
[126][75]. Although several efforts have been dedicated to understanding the precise mechanisms prompting αSyn accumulation, propagation and neurodegeneration in MSA, they still remain elusive. However, the release of αSyn by both neuronal and glial cells together with the stepwise spreading of αSyn pathology favors the contribution of the extracellular αSyn as a potential pathogenic ‘prion-like’ agent
[127][76] and the exosomes as a possible mediating pathway of αSyn propagation. Notwithstanding, the significant achievements and existing literature regarding the association of extracellular vesicles, and exosomes in particular, with the progression of PD pathology, it seems that
wresearche
rs are only at the tip of the iceberg concerning the implication of these nanovesicles in MSA-related pathology.
In an attempt to identify blood-based biomarkers to facilitate the differential diagnosis between PD, MSA and progressive supranuclear palsy (PSP) and to monitor disease progression, an initial study assessed the abundance of brain-derived exosomes of neuronal, astroglial or oligodendroglial origin isolated from blood plasma
[128][77]. From this analysis, it was reported that the plasma levels of neuronal-derived exosomes are statistically significantly lower in MSA patients compared to PD patients. In another study, whereby the αSyn cargo within serum neuronal exosomes has been estimated, it has become evident that exosomal αSyn in subjects affected by MSA was unchanged compared to control subjects and concomitantly was two-fold less than that measured in PD patients
[129,130][78][79]. Both the unaltered amounts of neuron-derived exosomes and the unchanged αSyn levels within these specific nanovesicles in the blood of MSA patients can be associated with the oligodendroglial nature of the disease. Therefore, a subsequent study attempted to untangle the role of oligodendroglial-derived exosomes as a key participant in MSA pathogenesis. More specifically, it was shown that the concentration of oligodendroglial-derived exosomes and αSyn cargo in these exosomes in the plasma of MSA patients were lower as compared to the respective exosomal concentrations in the plasma of age- and sex-matched healthy controls
[15]. This difference in αSyn content though was not associated with packaging less of αSyn into MSA-related exosomes since the average αSyn concentration per extracellular vesicle remained unaltered. In accordance with the patient derived data, in vivo data from a PLP-αSyn transgenic mouse model verified the reduced secretion of CNPase-positive oligodendroglial-derived exosomes in the plasma of these mice compared to wild-type mice or the PD-relevant Prnp-αSyn A53T mice. Interestingly, this difference became more apparent as the animals grow up, thus providing a link between age and alterations in exosomal release
[15]. Moreover, in the same study it was shown that the release of oligodendroglial-derived exosomes was declined in oligodendrocytes overexpressing human αSyn, or exposed to oligomeric αSyn or treated with GCI-derived αSyn aggregates from MSA patients, as compared to control cells
[15]. The mechanism involved was likely related, at least in part, to an αSyn-mediated interference in the interaction between syntaxin 4 and VAMP2, leading to the dysfunction of the SNARE complex and to the impairment of MVB docking into the plasma membrane and the subsequent exosome release
[15].
In contrast to the previous studies, another report presented conflicting data regarding the exosome-associated αSyn cargo derived from the blood serum of MSA patients
[131][80]. In more detail, they identified elevated concentrations of αSyn within both neuronal- and oligodendroglial-derived exosomes in the blood serum of MSA patients compared to the blood serum of healthy controls and PD patients. Given that MSA patients exhibit higher levels of αSyn in exosomes derived from oligodendrocytes than neurons, whereas the opposite is observed in PD patients, this
study combined the average total levels in each cell type into one biomarker and surmised that the ratio of αSyn concentration in oligodendroglial exosomes compared to neuronal exosomes enables to distinguish between the two synucleinopathies, with high sensitivity and specificity
[131][80]. In an attempt to develop a diagnostic model based on plasma-derived subpopulations of EVs in PD, MSA and atypical parkinsonism with tauopathy, a different study reported that immune surface markers in plasma-originated EVs were differentially expressed among PD, MSA patients and healthy controls and thus they could also be utilized as reliable biomarkers for diagnostic purposes
[132][81]. Finally, microglia/macrophage-derived EVs were also proposed to play a key role in MSA pathogenesis and specifically in αSyn spreading. In more detail, it has been shown that such exosomes isolated from the CSF of MSA patients contained elevated levels of αSyn species (both oligomeric and fibrillar) when compared to control samples and it is also noteworthy that they were able to induce αSyn aggregation in recipient cortical neurons
[13].