Oligodendrocytes Seem to Imprint Specific Alter Ego on α-Synuclein

You can take the protein out of the oligodendrocyte, but you can’t take the oligodendrocyte out of the protein. In the May 9 Nature, researchers led by Virginia Lee at the University of Pennsylvania, Philadelphia, make the claim that cell type determines which sort of toxic strain α-synuclein will become. The scientists compared neuronal α-synuclein aggregates found in Lewy body diseases such as Parkinson’s to those in the glial cytoplasmic inclusions (GCIs) found in multiple system atrophy (MSA). They confirmed that they represent two distinct conformations of α-synuclein, with the GCI strain being much more potent at coaxing normal α-synuclein to adopt a malevolent alter ego. While neurons and oligodendrocytes could take up either strain equally well, oligodendrocytes turned both Lewy body and synthetic α-synuclein into the GCI strain. On the other hand, GCI α-synuclein passaged through neurons maintained its identity. The findings suggest that the oligodendrocyte milieu somehow gives rise to the aggressive pathology found in MSA. MSA worsens more rapidly than PD.

Bad Seeds? The oligodendrocyte milieu gives rise to the aggressive α-synuclein pathology found in multiple system atrophy according to a new study. [Courtesy of R. Douglas Fields, NIH.]

“This is extremely exciting, because it takes this notion of ‘strain’ to a new level,” Patrik Brundin of the Van Andel Research Institute in Grand Rapids, Michigan, told Alzforum. He said the data offer surprising answers to key questions about strain formation. They also raise a new one—what is it about oligodendrocytes that makes α-synuclein assemblies shift into a particularly pathogenic conformation?

Others agreed it will be important to find the answer. “Identification of the cell-specific factors that impel α-synuclein to aggregate into the MSA strain could reveal ways to treat this debilitating and ultimately fatal brain disorder,” wrote Lary Walker of Emory University, Atlanta, in an accompanying editorial.

Previous work from Lee’s and other labs had shown the existence of distinct α-synuclein strains (see Jul 2013 newsBousset et al., 2013Peelaerts et al., 2015). Stanley Prusiner of the University of California, San Francisco, reported that aggregates from MSA patients behave unlike those from PD patients, with only the former seeding aggregation in transgenic mice, and both differing from synthetic α-synuclein fibrils (Prusiner et al., 2015Woerman et al., 2018).

Lee and colleagues dug deeper to compare GCI to Lewy body aggregates. First author Chao Peng isolated insoluble α-synuclein from postmortem brain samples from patients with MSA, PD, or dementia with Lewy bodies. The MSA strain was less phosphorylated at serine 129, better resisted proteinase K digestion, and bound more strongly to the antibody Syn7015, which primarily recognizes synthetic α-synuclein but in this study served to distinguish between GCI and LB forms of α-syn aggregate. The GCI material more aggressively seeded aggregation, with 30 picograms triggering as much α-synuclein fibrillization in oligodendrocyte cultures as did 30 nanograms of LB α-synuclein. GCI synuclein was also 1,000 times more potent than synthetic α-synuclein fibrils. The authors could not obtain enough purified aggregate to structurally characterize the two strains, although they plan to try.

Because GCIs are found in oligodendrocytes in people with MSA, the authors hypothesized that these deposits would preferentially seed aggregation in this cell type, while Lewy body aggregates might preferentially seed aggregation in neurons. To their surprise, however, the two types of aggregate showed no cell-type preference. In neurons, GCI strains triggered α-synuclein aggregation 1,000 times more potently than did LB strains, just as they did in oligodendrocytes, and the same thing happened in a human kidney cell line that expressed α-synuclein. All told, it seemed that the strain of aggregated α-synuclein in GCIs was more virulent and could propagate itself through any cell type.

Would this hold up in vivo? Peng and colleagues injected human α-synuclein taken postmortem from GCIs or Lewy bodies into wild-type mouse striatum. Three months later, LB α-synuclein had triggered no aggregates, while the GCI material had seeded numerous neuronal inclusions, though curiously none in oligodendrocytes, the type of cell in which the strain originated. Lee and colleagues suggest this might be because human and mouse oligodendrocytes normally express very little α-synuclein (Koga and Dickson, 2018). To confirm that the GCI strain can propagate faithfully through neurons, the authors added human GCI extracts to mouse primary neuron cultures. Newly seeded aggregates maintained the properties of GCI α-synuclein with regard to seeding and proteinase K digestion, even when passaged twice more through new neuronal cultures.

Different Strains. α-Synuclein deposits from MSA patients (middle) induce far more aggregated α-synuclein (red) in cultured oligodendrocytes than do synthetic fibrils (left) or synuclein from Lewy bodies (right). Nuclei stained blue. [Courtesy of Peng et al., 2018, Nature.]

Why then, does the GCI strain arise in oligodendrocytes to begin with? Peng and colleagues hypothesized that something about the intracellular environment might bring it about. To test this, they crossed α-synuclein knockout mice with M2 mice that overexpressed human α-synuclein only in oligodendrocytes. They injected GCI or LB α-synuclein into the thalami of the offspring at three months of age. Both types of aggregate seeded oligodendrocyte inclusions, although GCI material worked faster, with inclusions peaking by three months, versus six months for LB material. Intriguingly, the inclusions seeded by LB α-synuclein reacted with Syn7015, hinting they had assumed a GCI conformation. In other words, when expression of synuclein was artificially high, oligodendrocytes apparently converted LB aggregates into GCI aggregates. Both types of starting material seeded pathology in identical subcellular localization.

To gather more hints that oligodendrocytes force α-synuclein aggregates to assume a more virulent conformation, the authors turned again to cell cultures. They added synthetic α-synuclein fibrils, which are only weakly pathogenic, to cultured oligodendrocytes, neurons, or the kidney cell line. Again, the aggregates that formed in oligodendrocytes were highly pathogenic in seeding assays, suggesting they had adopted the GCI conformation. “To me, this was the most provocative experiment. Oligodendrocytes induce this potent conformation of synuclein pathology,” Lee said.

What about oligodendrocytes gives rise to this toxic strain? As a first step toward dissecting this, Peng and colleagues mixed synthetic α-synuclein fibrils with lysates from oligodendrocytes or neurons. The oligodendrocyte lysates stimulated formation of GCI-like material, suggesting that soluble factors were at play. The authors are now fractionating the oligodendrocyte lysates to narrow down the factors responsible. Lee noted that protease treatment blocks the effect, suggesting the crucial factor might be a protein.

Considering alternative explanations, Ole Isacson of Harvard Medical School’s McLean Hospital, Belmont, Massachusetts, wondered if lipids might promote the aggressive GCI strain. He noted that oligodendrocytes are particularly rich in lipid content. “It’s evident from our clinical brain samples that lipid-dependent cellular stresses are likely pathological triggers in PD and maybe MSA,” Isacson told Alzforum. Researchers in Switzerland and the Netherlands recently reported that Lewy bodies consist mostly of undigested vesicles, membranes, and other lipids, far outweighing the amount of α-synuclein and other proteins. A preprint of that paper has been posted to BioRxiv.

Isacson also noted that it remains to be seen how the data relate to disease, since the α-synuclein phosphorylation and misfolding used as readouts for the different strains may not mirror neurodegeneration seen in PD and MSA. Cells may even clear these strains through normal recycling processes. “They found that in specific mice directly injected into the brain with GCI synuclein strains, there was less misfolded α-synuclein present six months after injection than at three months after injection,” Isacson wrote. “Our data suggest that in the absence of lipid, metabolic, and autophagic problems, the abnormal proteins are eventually cleared by the cell.”

Another key question is how GCIs form in oligodendrocytes in the first place, given the low levels of α-synuclein in these cells. Perhaps some environmental insult such as infection or toxin exposure upregulates α-synuclein expression, Brundin suggested. A majority of the risk for PD and MSA appears to be environmental, rather than genetic, Brundin added. Some researchers have found evidence that cells release α-synuclein in response to infections, and this helps the animal survive (Tomlinson et al., 2017Stolzenberg et al., 2017).

Beyond these mechanistic questions, researchers see therapeutic implications. “Immunotherapies may have to be specific to the disease; one that works for PD may not work for MSA,” Brundin suggested. He wondered if clinical subtypes of Parkinson’s might also be characterized by distinct protein strains. For example, mutations in the gene encoding glucocerebrosidase typically cause a fast-progressing form of PD. Finding the answers to these questions could lead to more precise targeting of therapies to individual patients, Brundin said.

Featured Papers

Peng C, Gathagan RJ, Covell DJ, Medellin C, Stieber A, Robinson JL, Zhang B, Pitkin RM, Olufemi MF, Luk KC, Trojanowski JQ, Lee VM. Cellular milieu imparts distinct pathological α-synuclein strains in α-synucleinopathies. Nature. 2018 May 9; PubMed.

Walker LC. Sabotage by the brain’s supporting cells helps fuel neurodegeneration. Nature, May 9, 2018.


Bousset L, Pieri L, Ruiz-Arlandis G, Gath J, Jensen PH, Habenstein B, Madiona K, Olieric V, Böckmann A, Meier BH, Melki R. Structural and functional characterization of two alpha-synuclein strains. Nat Commun. 2013;4:2575. PubMed.

Woerman AL, Kazmi SA, Patel S, Aoyagi A, Oehler A, Widjaja K, Mordes DA, Olson SH, Prusiner SB. Familial Parkinson’s point mutation abolishes multiple system atrophy prion replication. Proc Natl Acad Sci U S A. 2018 Jan 9;115(2):409-414. Epub 2017 Dec 26 PubMed.

Koga S, Dickson DW. Recent advances in neuropathology, biomarkers and therapeutic approach of multiple system atrophy. J Neurol Neurosurg Psychiatry. 2018 Feb;89(2):175-184. Epub 2017 Aug 31 PubMed.

To view commentaries, primary articles and linked stories, go to the original posting on Alzforum.org here.

Copyright © 1996–2018 Biomedical Research Forum, LLC. All Rights Reserved.

Share this: