Search

α-Synucleinopathies are spreading neurodegenerative disorders characterized by the intracellular accumulation of insoluble aggregates populated by α-Synuclein (α-Syn) fibrils. In Parkinson's disease (PD) and dementia with Lewy bodies, intraneuronal α-Syn aggregates are referred to as Lewy bodies in the somata and as Lewy neurites in the neuronal processes. In multiple system atrophy (MSA) α-Syn aggregates are also found within mature oligodendrocytes (OLs) where they form Glial Cytoplasmic Inclusions (GCIs). However, the origin of GCIs remains enigmatic: (i) mature OLs do not express α-Syn, precluding the seeding and the buildup of inclusions and (ii) the artificial overexpression of α-Syn in OLs of transgenic mice results in a burden of soluble phosphorylated α-Syn but fails to form α-Syn fibrils. In contrast, mass spectrometry of α-Syn fibrillar aggregates from MSA patients points to the neuronal origin of the proteins intimately associated with the fibrils within the GCIs. This suggests that GCIs are preassembled in neurons and only secondarily incorporated into OLs. Interestingly, we recently isolated a synthetic human α-Syn fibril strain (1B fibrils) capable of seeding a type of neuronal inclusion observed early and specifically during MSA. Our goal was thus to investigate whether the neuronal α-Syn pathology seeded by 1B fibrils could eventually be transmitted to OLs to form GCIs in vivo. After confirming that mature OLs did not express α-Syn to detectable levels in the adult mouse brain, a series of mice received unilateral intra-striatal injections of 1B fibrils. The resulting α-Syn pathology was visualized using phospho-S129 α-Syn immunoreactivity (pSyn). We found that even though 1B fibrils were injected unilaterally, many pSyn-positive neuronal somas were present in layer V of the contralateral perirhinal cortex after 6 weeks. This suggested a fast retrograde spread of the pathology along the axons of crossing cortico-striatal neurons. We thus scrutinized the posterior limb of the anterior commissure, i.e., the myelinated interhemispheric tract containing the axons of these neurons: we indeed observed numerous pSyn-positive linear Lewy Neurites oriented parallel to the commissural axis, corresponding to axonal segments filled with aggregated α-Syn, with no obvious signs of OL α-Syn pathology at this stage. After 6 months however, the commissural Lewy neurites were no longer parallel but fragmented, curled up, sometimes squeezed in-between two consecutive OLs in interfascicular strands, or even engulfed inside OL perikarya, thus forming GCIs. We conclude that the 1B fibril strain can rapidly induce an α-Syn pathology typical of MSA in mice, in which the appearance of GCIs results from the pruning of diseased axonal segments containing aggregated α-Syn.

In 1957, Lionel Penrose built the first man-made self-replicating mechanical device and illustrated its function in a series of machine prototypes, prefiguring our current view of the genesis and the proliferation of amyloid fibrils. He invented and demonstrated, with the help of his son Roger, the concepts that decades later, would become the fundamentals of prion and prion-like neurobiology: nucleation, seeding and conformational templating of monomers, linear polymer elongation, fragmentation, and spread. He published his premonitory discovery in a movie he publicly presented at only two conferences in 1958, a movie we thus reproduce here. By making a 30-year-jump in the early 90's, we evoke the studies performed by Peter Lansbury and his group in which α-Synuclein (α-Syn) was for the first time (i) compared to a prion; (ii) shown to contain a fibrillization-prone domain capable of seeding its own assembly into fibrils; (iii) identified as an intrinsically disordered protein (IDP), and which, in the early 2000s, (iv) was described by one of us as a protein chameleon. We use these temporally distant breakthroughs to propose that the combination of the chameleon nature of α-Syn with the rigid gear of the Penrose machine is sufficient to account for a phenomenon that is of current interest: the emergence and the spread of a variety of α-Syn fibril strains in α-Synucleinopathies.

The distinct neuropathological features of the different α-Synucleinopathies, as well as the diversity of the α-Synuclein (α-Syn) intracellular inclusion bodies observed in post mortem brain sections, are thought to reflect the strain diversity characterizing invasive α-Syn amyloids. However, this "one strain, one disease" view is still hypothetical, and to date, a possible disease-specific contribution of non-amyloid factors has not been ruled out. In Multiple System Atrophy (MSA), the buildup of α-Syn inclusions in oligodendrocytes seems to result from the terminal storage of α-Syn amyloid aggregates first pre-assembled in neurons. This assembly occurs at the level of neuronal cytoplasmic inclusions, and even earlier, within neuronal intranuclear inclusions (NIIs). Intriguingly, α-Syn NIIs are never observed in α-Synucleinopathies other than MSA, suggesting that these inclusions originate (i) from the unique molecular properties of the α-Syn fibril strains encountered in this disease, or alternatively, (ii) from other factors specifically dysregulated in MSA and driving the intranuclear fibrillization of α-Syn. We report the isolation and structural characterization of a synthetic human α-Syn fibril strain uniquely capable of seeding α-Syn fibrillization inside the nuclear compartment. In primary mouse cortical neurons, this strain provokes the buildup of NIIs with a remarkable morphology reminiscent of cat's eye marbles (see video abstract). These α-Syn inclusions form giant patterns made of one, two, or three lentiform beams that span the whole intranuclear volume, pushing apart the chromatin. The input fibrils are no longer detectable inside the NIIs, where they become dominated by the aggregation of endogenous α-Syn. In addition to its phosphorylation at S129, α-Syn forming the NIIs acquires an epitope antibody reactivity profile that indicates its organization into fibrils, and is associated with the classical markers of α-Syn pathology p62 and ubiquitin. NIIs are also observed in vivo after intracerebral injection of the fibril strain in mice. Our data thus show that the ability to seed NIIs is a strain property that is integrally encoded in the fibril supramolecular architecture. Upstream alterations of cellular mechanisms are not required. In contrast to the lentiform TDP-43 NIIs, which are observed in certain frontotemporal dementias and which are conditional upon GRN or VCP mutations, our data support the hypothesis that the presence of α-Syn NIIs in MSA is instead purely amyloid-strain-dependent.