Basal Ganglia: Network Dynamics in Brain Function and Dysfunction

The basal ganglia (BG) play a crucial role in motor control and multiple cognitive functions. BG dysfunction can lead to brain diseases, e.g. Parkinson’s (PD) and Huntington’s disease. Independent of their etiology, most BG-related brain disorders manifest themselves in aberrant neuronal activity in the BG. Restoration of a healthy activity state (e.g. by deep brain stimulation - DBS) often ameliorates disease symptoms. Thus, many BG-related brain disorders can be considered a disease of brain dynamics and, hence, it was only natural for BCF researchers to focus on the BG. It was also an opportune time, because several new developments had just started challenging the simplified description of the BG as a feedforward network.

Since 2010, we have elucidated the role of BG structure in shaping the dynamics, developed multi-scale models of aberrant activity associated with PD and possible stimulation protocols to restore healthy BG activity and function as summarized below:

Modeling beta-band oscillations:

We proposed a model to explain both, the emergence of persistent beta-band oscillations in PD (Kumar et al. 2011) and transient beta-band oscillations in decision-making in healthy animals (Mirzaei et al. 2017). In this model, activity of striatal neurons projecting to the globus pallidus externa (GPe) controls the power of beta-band oscillations in the BG (Kumar et al. 2010). Next to explaining many properties of beta-band oscillations, the model also revealed how increasing spiking bursting in GPe could exacerbate oscillations (Sahasranamam et al. in prep.).

Closed-loop DBS protocol:

Extending our model, we proposed the first closed-loop DBS protocol to not just quench beta-band oscillations but also to restore the healthy state of network computation in a spiking neuronal network model (Vlachos et al. 2016).

Role of striatal connectivity in shaping network dynamics:

We proposed a novel functional consequence of the structure of cortico-striatal projections (Yim et al. 2011). Next, we found that the unequal connectivity between D1- and D2-type dopamine receptor expressing striatal neurons gives rise to a ‘decision transition threshold (DTT; Bahuguna et al. 2015). DTT controls the balance of direct and indirect pathways, providing a mechanistic understanding of many BG-related brain disorders. Finally, we showed that the emergence of spatially localized cell assemblies requires the distance-dependent connectivity among striatal neurons to follow a non-monotonic profile (Spreizer et al. 2017).

Spatio-temporal structure of dopamine:

By modeling the diffusion of dopamine in a medium crowded with neurons and glia, we found that degeneration of dopaminergic axons (as in PD) should only affect dopamine transients and not the average levels (Hunger et al. in prep.).

Discovery of new changes in the BG during PD:

In collaboration with experimentalists we identified two novel changes associated with PD: (1) dopamine depletion increases the variability of spiking activity in the globus pallidus interna (GPi) - the output of the basal ganglia (Spreizer et al. 2017b; collaboration with Atsushi Nambu, NIPS, Okazaki, Japan). (2) In healthy animals, cortico-striatal synapses on D1-SPNs are stronger than those on D2- SPNs, whereas in PD there is no such difference (Filipovic et al. 2017; collaboration with Gilad Silberberg, Karolinska Institute, Stockholm, Sweden).

Figure:

Platzhalterbild