The sulfation code for propagation of neurodegeneration

Prion-like propagation of protein aggregates is thought to be an essential feature in many neurodegenerative diseases, but the mechanisms underlying transcellular transfer of protein aggregates remain unclear. Stopschinski et al. now demonstrate that the cellular uptake of tau, Aβ, and α-synuclein aggregates mediated by heparan sulfate proteoglycans (HSPGs) varies with distinct glycosaminoglycan chain length and sulfation patterns. The results help us to understand how different protein aggregates propagate, leading to distinct neurodegenerative pathologies.

Prion-like propagation of protein aggregates is thought to be an essential feature in many neurodegenerative diseases, but the mechanisms underlying transcellular transfer of protein aggregates remain unclear. Stopschinski et al. now demonstrate that the cellular uptake of tau, A␤, and ␣-synuclein aggregates mediated by heparan sulfate proteoglycans (HSPGs) varies with distinct glycosaminoglycan chain length and sulfation patterns. The results help us to understand how different protein aggregates propagate, leading to distinct neurodegenerative pathologies.
Prion diseases are a group of fatal transmissible neurodegenerative diseases characterized by the accumulation of an abnormal form of prion protein (PrP Sc ) in the central nervous system (CNS). 2 Prion diseases are transmissible interindividually, crossing even species barriers, and propagate intraindividually. The transmission is mediated by transfer of PrP Sc , which functions as a template for the conversion of normal cellular forms of prion protein (PrP C ) to PrP Sc .
Similar to prion diseases, many neurodegenerative diseases are also characterized by the accumulation of abnormal proteins in the CNS including ␤-amyloid protein (A␤) and tau in Alzheimer's disease (AD), tau in non-AD tauopathies, ␣-synuclein in Lewy body diseases such as Parkinson's disease and dementia with Lewy bodies, and transactive response DNAbinding protein 43 kDa (TDP-43) in frontotemporal dementia and amyotrophic lateral sclerosis (1-3). Increasing evidence from experimental studies has indicated that misfolded aggregated proteins in these diseases have prion-like properties: They propagate through neuronal networks or other pathways in the CNS of individuals and also cause interindividual transmission of these diseases. In human studies, neurodegeneration seems to progress along the neuronal networks by propagation of protein aggregates in the CNS, although evidence for human-to-human transmission is still limited (4,5).
For intercellular propagation of intracellular protein aggregates, release of aggregates from donor cells and uptake of aggregates by recipient cells are essential steps, followed by intracellular seeding in the recipient cells. In transcellular transfer of protein aggregates, multiple mechanisms have been proposed: 1) extracellular vesicles like exosomes and ectosomes reaching the cytoplasm of recipient cells by fusion, 2) free protein aggregates exocytosed from donor cells taken up by recipient cells with receptor-or nonreceptor-mediated endocytosis/macropinocytosis, and 3) transfer through nanotubes, although it remains unclear which is the main mechanism for transcellular transfer of certain protein aggregates in certain areas of the CNS.
Diamond and colleagues (6) have focused on heparan sulfate proteoglycans (HSPGs) on the cell surface as receptors for uptake of protein aggregates such as A␤, tau, and ␣-synuclein. They previously observed that cellular uptake and consequent intracellular seeding of tau and ␣-synuclein fibrils require HSPGs, similar to results from another study (7). Inhibition of the interaction with HSPGs blocked transcellular aggregate propagation, indicating that the interaction between HSPG and protein aggregates may be an essential step for propagation of neurodegeneration induced by protein aggregates. However, as the properties of HSPGs vary with different glycosaminoglycan (GAG) compositions, it is unclear which properties of GAGs are specifically required for binding of different protein aggregates to mediate uptake and intracellular seeding. Now, Diamond and colleagues (8) report the specificity for interactions of the aggregates with HSPGs. They determined the size and sulfation requirements of aggregate binding to GAGs by measuring direct binding to modified heparins in carbohydrate microarrays and by using competition studies with heparin derivatives in cell-based assays. Specifically, they observed that the binding of tau aggregates required a minimal length and precise GAG architecture with defined sulfate moieties in the N-and 6-Opositions leading to seeding activity inside cells (as reported using biosensor assays) (Fig. 1), whereas binding of A␤ and ␣-synuclein aggregates displayed a more complex behavior. Huntingtin fibrils exhibited no heparin binding. Individually knocking out the major genes of the HSPG pathway in HEK293T cells corroborated the microarray and pharmacological data that the uptake of tau is mediated by N-and 6-O-sulfation of HSPGs.
The results by Diamond and colleagues reveal that specific GAG chain length and sulfation patterns are necessary for cel- lular uptake of tau versus ␣-synuclein and A␤ aggregates. These data inspire yet more questions concerning the full role of HSPGs in transcellular propagation of protein aggregates in neurodegenerative diseases. For example, the results for binding of ␣-synuclein aggregates to HSPGs were variable among different experimental approaches, suggesting that interactions between protein aggregates and HSPGs are very complex. Further exploration of other molecular determinants that govern the binding and internalization of each aggregated protein would be interesting. Second, the authors used fibrils of tau, ␣-synuclein, A␤, and huntingtin in this study. However, smaller nonfibrillar aggregates, such as soluble oligomers, could be important for cell-to-cell propagation. It was reported for ␣-synuclein that internalization of amyloid fibrils depends on heparin sulfate, whereas that of smaller non-amyloid oligomers does not (7). Thus, binding of smaller oligomeric forms of tau and other protein aggregates to HSPGs should be investigated.
In addition, fibrils formed in vitro from synthetic or recombinant proteins consist of a mixture of different fibrillary structures, such as spiral and straight fibrils (9). Different fibrillar structures of the same protein may represent different "strains" associated with peculiar patterns of neuropathological lesions. These different fibrillary structures of the same protein may require different HSPG structures for cellular uptake, a possibility that should be examined. Finally, it would be fascinating to evaluate compositions of HSPGs on the cell surface of the brain and peripheral tissues that actually could be involved in propagation of protein aggregates in vivo. Neuronal subtypes that express HSPGs with distinct GAG chain length and sulfation patterns specific for protein aggregates would be preferentially affected in neurodegenerative disorders due to high levels of uptake and intracellular aggregation. Such studies help us to understand how different "strains" of the same protein aggregates, as well as different protein aggregates, produce distinct progression patterns of pathologies in neurodegenerative diseases. Treatment with heparan sulfate mimetics has shown beneficial effects in animal models of prion diseases or neurodegenerative diseases with prion-like propagation of protein aggregates; however, clinical trials have been unsuccessful so far (10). Further understanding of molecular mechanisms underlying HSPG-mediated cellular uptake of protein aggregates is essential for the development of HSPG-targeting therapies to inhibit progression of neurodegenerative diseases.