Highlights
- •Prionoid protein propagation is a common mechanism of neurodegenerative diseases
- •Neuron-to-neuron propagation of aggregated disease proteins has been assumed
- •We reveal propagation of the monomer disease protein via brain lymphatic system
- •The brain lymphatic propagation of monomer is a target of future therapeutics
Summary
Prion-like protein propagation is considered a common pathogenic mechanism in neurodegenerative diseases. Here we investigate the in vivo propagation pattern and aggregation state of mutant α-synuclein by injecting adeno-associated viral (AAV)-α-synuclein-A53T-EGFP into the mouse olfactory cortex. Comparison of aggregation states in various brain regions at multiple time points after injection using western blot analyses shows that the monomeric state of the mutant/misfolded protein propagates to remote brain regions by 2 weeks and that the propagated proteins aggregate in situ after being incorporated into neurons. Moreover, injection of Alexa 488-labeled α-synuclein-A53T confirms the monomeric propagation at 2 weeks. Super-resolution microscopy shows that both α-synuclein-A53T proteins propagate via the lymphatic system, penetrate perineuronal nets, and reach the surface of neurons. Electron microscopy shows that the propagated mutant/misfolded monomer forms fibrils characteristic of Parkinson’s disease after its incorporation into neurons. These findings suggest a mode of propagation different from that of aggregate-dependent propagation.
Introduction
Prion diseases, which include Creutzfeldt-Jacob disease and bovine spongiform encephalopathy, are neurodegenerative disorders mediated by transmissible proteins whose misfolded structures change the normal structures of prion proteins to misfolded and aggregation-prone structures. Prion diseases progress rapidly, whereas the incubation period from infection to symptomatic onset is extremely long, except in the case of Creutzfeldt-Jacob disease derived from bovine spongiform encephalopathy (CDC prion diseases: https://www.cdc.gov/prions/index.html). The similar propagation of prionoids has been suggested in neurodegenerative diseases, which could depend on cell death, microglia, exo-endocytosis, and trans-synaptic mechanisms.
In Parkinson’s disease (PD), gut-to-brain transmission via the vagus nerve has been suggested by staging patients, while secondary propagation from the olfactory bulb to the limbic system (including the entorhinal cortex, amygdala, and hippocampus) can trigger dementia with Lewy bodies. Intravenous injection of synthetic α-synuclein leads to cell-to-cell transmission and causes PD-like pathology in non-transgenic mice. Injections of α-synuclein with different aggregation and conformational states cause different pathologies in mouse brains. In particular, processes related to cell-to-cell transmission, ectosomes, exosomes, receptor-mediated uptake, endocytosis, macropinocytosis, and nanotubes are implicated in such pathologies.
A popular experimental technique for analyzing prionoid propagation is to inject protofibrils or full fibrils, generated in vitro or extracted from human patient brains, into animal brains and then observe their propagation and any accompanying effects. However, such experiments do not directly answer which states of prionoid proteins (i.e., monomer, oligomer, protofibril, or fibril) are propagated or the underlying mechanism in the brain in vivo.
To determine which protein state is propagated, we propose examining the states of proteins that propagate after their synthesis de novo in neurons, instead of examining the states of proteins that propagate from exogenously injected pre-synthesized fibrils. In this study, local injection of a very small dose of adeno-associated viral (AAV) vector revealed the pathway of propagation of disease-associated α-synuclein carrying the A53T mutation (α-synuclein-A53T). Our results suggest that a part of α-synuclein is transported in its monomeric state to remote brain regions faster than previously reported via the brain lymphatic system. The aggregates that form after monomer propagation still possess the ribbon fibrils characteristic of PD, suggesting that the mutant/misfolded monomer might form different aggregation species after propagation.
Results
Propagation of endogenous α-synuclein in the brain
We injected a very small amount of AAV-α-synuclein-A53T-EGFP or AAV-EGFP (3.5 × 107 vector genomes [vg] in 1 μL) into the orbitofrontal cortex at the boundary between the lateral orbital cortex (LO) and frontal associated cortex (FrA) of 6-week-old C57BL/6J mice (Figure 1A). This region transfers the input from the olfactory bulb to the frontal cortex, thalamus, and other brain regions. Here we did not select a neuron-specific promoter for AAV vector because we did not want to exclude any mode of propagation, including propagation from any cell type in the olfactory pathway to neurons in remote brain regions. The EGFP signal was restricted to the injection site at 2 weeks post injection, and instead of increasing, decreased at the injected site at 12 months post injection (Figure 1B). The signal intensities were far lower than those observed in mice injected with a therapeutic dose of AAV-EGFP into the cerebellar surface (Figure 1C). At 12 months post injection, we made the brains transparent by the CUBIC1 method (Figure 1C) and observed them via light sheet microscopy (Figures 1D–1F) and confocal microscopy (Figures 1G–1M).
The observations revealed unexpected distribution patterns of the expressed protein (Figures 1D–1F and Video S1). First, the EGFP signals expanded to remote brain regions, both after AAV-α-synuclein-A53T-EGFP injection and after AAV-EGFP injection (Figure 1E), suggesting that proteins expressed in a narrow region propagated to remote brain regions non-specifically. Second, an image of α-synuclein-A53T-EGFP signal intensities subtracted for the EGFP signal intensities in each pixel in the same horizontal section (see STAR Methods for details) (Figures 1E and 1F) suggested high signals in some brain regions that might be attributed to specific α-synuclein-A53T propagation. To evaluate the second point more precisely, we injected AAV-mCherry and AAV-α-synuclein-A53T-EGFP simultaneously into the same mouse and compared mCherry signal intensities and EGFP signal intensities in a horizontal section after equalizing correction of their total signal intensities (Figure 1G). The merged image of mCherry and EGFP signals revealed that brain regions including the thalamic nuclei, striatum, SN, brain stem nuclei, and hippocampus CA2/3 were green dominant (Figure 1F). The result was further confirmed by quantification of the two color signals in different brain regions (Figure 1H)…
