tcwhalen_phd_ni_2021.pdf (2.16 MB)

Delta Oscillations: Prevalence, Propagation, and Relation to Motor Dysfunction in Mouse Models of Parkinson’s Disease

Download (2.16 MB)
thesis
posted on 07.05.2021, 20:04 by Timothy Whalen
Pathological neural oscillations, particularly in the beta and delta bands, are hallmarks of dysfunction in the basal ganglia (BG) of patients with Parkinson’s disease (PD). While Parkinsonian beta oscillations have received more attention than delta oscillations in the scientific literature, it remains unclear how these oscillations emerge, propagate through the brain, and relate to motor symptoms in PD. Animal models of PD have been a valuable tool for studying these oscillations, but the oscillatory landscape of awake, behaving mice, a common animal model for the study of PD, has not been investigated.
Here, we record from the substantia nigra pars reticulata (SNr), the primary output nucleus of the mouse BG, and other BG nuclei in the dopamine depleted (DD) mouse model of PD. Using a novel signal processing method to distinguish oscillations from neural noise, we establish that delta, but not beta, oscillations are present in single neural units throughout the BG in dopamine depletion, and that these oscillations arise due to insufficient activation of D2 receptors. We also establish that the prevalence of delta oscillations in SNr neurons correlates with the overall level of motor dysfunction and dopamine loss and dynamically correlates with bouts of akinesia. These oscillations in SNr neurons lead DD-induced delta oscillations in motor cortex (M1), suggesting a subcortical basis for their generation, and their relationship to M1’s oscillations define a novel dichotomy of SNr into active-predicting (AP) and inactive-predicting (IP) subpopulations of neurons.
Next, we take a computational modeling approach to further investigate how these oscillations propagate in the brain and how the AP and IP subpopulations of SNr neurons arise. Using a realistic conductance-based model of SNr neurons, we test if delta oscillations in GPe neurons which project to our model SNr neurons are sufficient to replicate the SNr oscillations we observe in vivo. We demonstrate that a simple connection architecture, in which GPe and other SNr neurons compete for a limited number of synapses on each SNr neuron, is sufficient to generate AP and IP populations in SNr whose firing properties match experimental data. This model demonstrates how delta oscillations can effectively propagate through the basal ganglia despite neural noise.
Finally, we review how these results fit within and inform our understanding of neural oscillations and Parkinsonian motor dysfunction. We discuss and attempt to reconcile the disparities between observations in different animal models and human PD and explore potential mechanisms by which delta oscillations could cause Parkinsonian dysfunction. We close with a discussion of the future directions we envision for these topics and how they may inform new potential targets and treatments for PD.

History

Date

23/02/2021

Degree Type

Dissertation

Department

Neuroscience Institute

Degree Name

  • Doctor of Philosophy (PhD)

Advisor(s)

Aryn Gittis Jonathan Rubin

Usage metrics

Categories

Licence

Exports