Objective: To investigate the ability of a susceptible-infected-removed (SIR) agent-based model to predict the pathogenic spread of alpha-synuclein fibrils after injection in mice.

Background: Parkinson’s disease is characterized by the intracellular accumulation of alpha-synuclein-containing aggregates. These aggregates are thought to spread between cells in a prion-like mechanism. The use of computational modeling is one approach to understand the spreading mechanisms of misfolded proteins through the brain. We have developed an agent-based SIR model that accounts for connectivity-based propagation and variations in regional vulnerability. This model was shown to replicate the deformation pattern of atrophy found in patients with Parkinson’s disease [1]. Here we test the model’s ability to predict alpha-synuclein propagation in mice.

Method: Eighty-nine non-transgenic mice were injected with 5 micrograms of alpha-synuclein preformed fibrils in either the caudoputamen (n=38), the nucleus accumbens (n=24) or the hippocampus (n=27). Mice were sacrificed at two weeks, 1, 3, 6, 12, 18 or 24 months post-injection. Phospho-Ser129 (81A) pathology was quantified on sections covering the 426 regions from the Allen Mouse Brain Atlas. The SIR simulator was used to simulate the spread of alpha-synuclein pathology in the same 426 regions, considering the whole-brain connectivity matrix from the Allen Mouse Brain Connectivity Atlas and the local SNCA expression as model parameters. Spearman’s rank correlation coefficients were used to assess the peak fits between observed and simulated data.

Results: The SIR simulator significantly replicated the distribution of Lewy-like soma and neuritic pathology observed in every dataset. The fits were significant at each post-injection timepoint and reached 0.65-0.71 for the caudoputamen, 0.67-0.74 for the nucleus accumbens, and 0.44-0.61 for the hippocampus. The peak fits were similar when simulating retrograde or anterograde spread of alpha-synuclein.

Conclusion: The agent-based SIR model replicates the spread of alpha-synuclein in mouse models.  This supports the use of this model to understand the spreading mechanisms involved in neurodegenerative diseases and to test the effect of putative neuroprotective therapies in cost-effective ways.

References: [1] Zheng YQ, Zhang Y, Yau Y, Zeighami Y, Larcher K, Misic B, Dagher A. (2019). Local vulnerability and global connectivity jointly shape neurodegenerative disease propagation. PLoS Biol. 17(11):e3000495.

« Back to MDS Virtual Congress 2020

MDS Abstracts - https://www.mdsabstracts.org/abstract/an-agent-based-model-replicates-the-pathogenic-spread-of-alpha-synuclein-in-mice/