Does a prion-like mechanism play a major role in the apparent spread of α-synuclein pathology?

Parkinson's disease, the most common movement disorder, results in an insidious reduction for patients in quality of life and ability to function. A hallmark of Parkinson's disease is the brain accumulation of neuronal cytoplasmic inclusions comprised of the protein α-synuclein. The presence of α-synuclein brain aggregates is observed in several neurodegenerative diseases, including dementia with Lewy bodies and Lewy body variant of Alzheimer's disease. These disorders, as a group, are termed synucleinopathies. Mounting evidence indicates that α-synuclein amyloid pathology may spread during disease progression by a prion-like (self-templating alteration in protein conformation) mechanism. Clear in vitro and cell culture data demonstrate that amyloidogenic α-synuclein can readily induce the conversion of other α-synuclein molecules into this conformation. Some data from experimental mouse studies and autopsied brain analyses also are consistent with the notion that a self-promoting process of α-synuclein amyloid inclusion formation may lead to a progressive spread of disease in vivo. However, as pointed out in this review, there are alternative explanations and interpretations for these findings. Therefore, from a therapeutic perspective, it is critical to determine the relative importance and contribution of α-synuclein prionlike spread in disease before embarking on elaborate efforts to target this putative pathogenic mechanism.

α-synuclein (SNCA) gene were the fi rst genetic causes of PD identifi ed and this fi nding led to a disease paradigm shift toward the importance of the aberrant aggregation of this protein in the etiology of PD. To date, three missense mutations (A53T, A30P, and E46K) have been identifi ed and linked to PD or DLB [19][20][21]. In addition, short chromosomal duplications or trisomies containing the SCNA gene, plus relatively short fl anking regions on chromosome 4, were discovered in patients with PD or DLB [22,23], indicating that a 50% increase in the expression of α-synuclein is suffi cient to cause disease. Although the precise mechanism of toxicity of α-synuclein is still debated, most evidence indicates that some form of protein aggregation is involved [13,24].

Evidence for a prion/spreading mechanism of α-synuclein pathology
Th e cause of disease progression in PD and DLB has long been elusive, but recent fi ndings suggest that α-synuclein aggregation may proceed via a 'prion-like' mechanism that leads to a spreading of α-synuclein pathology. Th e term 'prion-like' is used because there is no evidence that α-synuclein aggregates are transmitted between individuals (that is, there is no evidence for infectability), but as reviewed below, α-synuclein aggregation can clearly seed and self-template in vitro and in cultured cells under conditions that favor the entry of preformed amyloid into cells. Furthermore, it may be able to propagate between cells.
It is well established that in vitro α-synuclein aggregation into amyloid is a nucleation-dependent process and can be greatly induced by the addition of a 'seed' or 'nucleus' of pre-aggregated α-synuclein [37,38]. Cellular studies have shown that the entry of a small amount of preformed α-synuclein fi brils using reagents that promote the entry of these seeds across the plasma membrane can very effi ciently induce the formation of large intracellular amyloid inclusions [39][40][41]. More recently, it was shown that the simple addition of extracellular α-synuclein fi brils to primary neurons can also induce the formation of intracellular α-synuclein inclusions [42]. Collectively, these studies suggest that αsynuclein amyloid formation may have the ability to spread between cells, although in a robust cellular model of seeded intracellular α-synuclein inclusion formation, it was not possible to observe transmission or propagation of α-synuclein amyloid between cells [40].
A prion-like spreading of α-synuclein pathology is consistent with the Braak staging of disease that appears to follow neuroanatomical pathways [43]. Th is staging is based on the semi-quantitative immunocytochemical analysis of brain α-synuclein pathology distribution in patients with PD and age-matched controls and indicates a temporal sequence or stages of ascending severity. Other studies also suggest that α-synuclein can spread in human brains and that it may even start in the peripheral nervous system, more specifi cally the enteric nervous system [44][45][46]. Th is has led to the speculation that a pathogen, perhaps a virus or agent that could alter αsynuclein conformation, may initiate disease at the periphery and propagate back to the brain. Th e in vivo propagation of α-synuclein is also suggested by the presence of LB formation in fetal dopaminergic neurons that were transplanted in the striatum of PD patients as attempted therapeutic interventions [47][48][49].
More recently, it was reported that intracerebral injection of extracts from sick A53T human α-synuclein trans genic mice (line M83) into younger healthy M83 transgenic mice could induce disease [50,51]. Th e M83 transgenic mice normally develop a late-onset (8 to 15 months of age) severe motor phenotype that leads to death and that results from the widespread formation of neuronal α-synuclein amyloidogenic inclusions [52]. Th e intracerebral injection of extracts from aff ected M83 mice resulted in an earlier presentation of both pathology and phenotype, and this induction was not observed in the younger mice injected with extracts generated from healthy mice [50,51]. Furthermore, brain injection of preformed recombinant α-synuclein fi brils can induce αsynuclein pathology that appears to spread from the injection site [50], suggesting that these α-synuclein species can initiate and perhaps lead to transmission of α-synuclein pathology. Nevertheless, it is also possible that the observed induced pathology from both types of injections could be the result of a focal brain insult that initiates alternatively proposed pathological cascades such as oxidative stress, excitotoxicity, and neuroinfl ammation [53] that could result in similar observations.

Is the apparent spread of α-synuclein due solely to intra-and intercellular templating of α-synuclein aggregation?
Although many lines of evidence support the notion that α-synuclein may propagate in vivo via a prion-like mechanism, several fi ndings are not completely consis tent with this model. For example, pathological assessment of A53T (line M83) and E46K (line M47) human α-synuclein transgenic mice demonstrated that α-synuclein inclusion formation in these mice is late-onset, relatively rapid, and likely synchronized, but there is a paucity of neuronal inclusion clustering that is incon sis tent with a spreading mechanism [54].
Although Lewy pathology was observed in some transplanted cells of several PD patients who received fetal mesencephalic grafts [47][48][49], this phenomenon was observed in only a minority of these patients, even though some patients without Lewy pathology in grafted cells survived for a similar period of time following the transplant surgery [47,53,55]. It is possible that transmission is a very slow or ineffi cient process or, as noted above, that other alternative mechanisms may also lead to the observed induction of Lewy pathology in the transplanted cells [53]. Furthermore, extensive reviews of large cohorts of autopsied patients with PD/DLB have revealed that, for a signifi cant percentage of patients, the distribu tion of α-synuclein pathology is not consistent with the Braak scheme of α-synuclein neuronal pathology spread [56][57][58].
Moreover, the progression of pathology via a prion-like mechanism is not consistent with what is observed in multiple system atrophy (MSA). MSA is an adult-onset neurodegenerative disease that is characterized by varying degrees of parkinsonian features, cerebellar ataxia, and autonomic dysfunction [59][60][61] but that is defi ned pathologically by the presence of glial cytoplasmic inclusions (GCIs) [62]. GCIs usually appear as fl ame or sickle inclusions in oligodendrocytes found throughout the white matter, but the greatest abundance of these inclusions occurs in the basal ganglia, the substantia nigra, the pontine nucleus, medulla, and cerebellum [63][64][65]. Like Lewy pathology, GCIs are comprised predominantly of polymerized amyloidogenic α-synuclein [63,66,67]. Whereas most pathological inclu sions in MSA are in oligodendrocytes, some α-synuclein protein aggregates can also be observed in the form of neuronal cytoplasmic inclusions, most of which are indistinguishable from LBs, and in neuritic processes, especially in the pontine nuclei and striatum [65,68,69]. α-Synuclein is expressed predominantly in neurons where it is localized to presynaptic terminals [70][71][72] and is expressed only at very low levels or levels below detectability in oligodendrocytes [73,74], and its expression is not increased in MSA [63,66,73]. Th erefore, it is unclear why oligodendro cytes are aff ected predominantly in MSA. More importantly, if a prion-like mechanism is the primary mechanism for the spread of α-synuclein pathology, why is it confi ned predominantly to cells that express low levels of α-synuclein in MSA? Why does it not effi ciently transmit to neurons? One possible mechanism that has been suggested to contribute to the spread of α-synuclein pathology is the release of amyloidogenic α-synuclein from dying cells. In MSA brains, 'ghost' GCIs where the cells have died are a common observation, suggesting that amyloidogenic α-synuclein from these cells should be readily available for the seeding in other neighboring cells.

Conclusions
Despite mounting evidence that a prion-like mechanism may contribute to the spread/progression of α-synuclein pathology in PD and DLB, several inconsistencies or alternative explanations [53,75] for the observed experimental fi ndings indicate that the pathogenic landscape is complex. Further experimental studies are needed to determine the relative importance and involvement of various pathogenic mechanisms, including self-templating resulting in altered protein conformation, neuroinfl am mation, oxidative stress, and excitotoxicity, in the apparent spread of α-synuclein pathology. Th is is especially important since extracellalar α-synuclein aggregates that would likely be involved in a prion-like spread of disease could lead to the design of novel therapeutic interventions such as conformation-specifi c monocloncal antibody therapy and regulation of specifi c extracellular proteases. Th e validity of pursuing this target will require a clearer understanding of both the pathological fi ndings and experimental paradigms that point in this direction. Nevertheless, the possibility of preventing PD/DLB disease progression by blocking brain prion-like amyloid spread is an attractive hypothesis.