Title: “Global Neural Rhythm Control by Local Neuromodulation”
Nancy Kopell, PhD – BU Math (Advisor)
Kamal Sen, PhD – BU BME (Chair)
John White, PhD – BU BME
Michelle McCarthy, PhD/MD – BU Math
Emery Brown, MD/PhD – MGH Anesthesia
Neural oscillations are a ubiquitous form of neural activity seen across scales and modalities. These neural rhythms correlate with diverse cognitive functions and brain states. One mechanism for changing the oscillatory dynamics of large neuronal populations is through neuromodulator activity. This dissertation explores the phenomenon of local neuromodulation of a distinct neuron type within a single brain nucleus exerting a powerful influence on global cortical rhythms.
One approach to investigate the impact of local circuits on global rhythms is through optogenetic techniques. My first project involves the statistical analysis of electrophysiological recordings of an optogenetically-mediated Parkinsonian phenotype. Empirical studies demonstrate that Parkinsonian motor deficits correlate with the emergence of exaggerated beta frequency (15-30 Hz) oscillations throughout the cortico-basal ganglia-thalamic network. However, the mechanism of these aberrant oscillatory dynamics is not well understood. A previous modeling study predicted that cholinergic neuromodulation of medium spiny neurons in the striatum of the basal ganglia is responsible for the pathologic beta rhythm. Here, this hypothesis was tested using selective optogenetic stimulation of striatal cholinergic interneurons in normal mice; stimulation robustly and reversibly amplified beta oscillations and Parkinsonian motor symptoms.
The modulation of global rhythms by local networks was further studied using computational modeling in the context of intrathalamic neuromodulation. While intrathalamic vasoactive intestinal peptide (VIP) is known to cause long-lasting excitation in vitro, its in vivo dynamical effects have not been reported. Here, biophysical computational models were used to elucidate the impact of VIP on thalamocortical dynamics during sleep and propofol general anesthesia. The modeling results suggest that VIP can form robust sleep spindle oscillations and control aspects of sleep architecture through a novel homeostatic mechanism. This homeostasis would be inhibited by general anesthesia, representing a new mechanism contributing to anesthetic-induced loss of consciousness.
While the previous two projects differed in their use of empirical versus theoretical methods, a challenge common to both domains is the difficulty in visualizing and analyzing large multi-dimensional datasets. A tool to mitigate these issues is introduced here: GIMBL-Vis is a Graphical Interactive Multi-dimensional extensiBLe Visualization toolbox for Matlab. This toolbox simplifies the process of exploring multi-dimensional data in Matlab by providing a graphical interface for visualization and analysis. Furthermore, it provides an extensible open platform for distributed development by the community.