Ivan Soltesz

Professor of Neurosurgery and Neurosciences
Neurosurgery Department
Stanford University School of Medicine
(November 24, 2015)

Organization and Control of Hippocampal Circuits

Neurons do not work alone. Individual neurons and interneurons are small parts of larger networks, and the carefully timed firing of neurons in a network form the basis of rhythms and functioning of the brain. The interneurons play their part by inhibiting the firing of other neurons, controlling the activity patterns and rhythms that control behavior. In his talk, Dr. Soltesz discussed his work examining the interneurons of the hippocampus. An understanding of how these networks are organized can and has led to methods of altering the firing patterns, increasing the possible ways for treating epilepsy and other disorders of network function.

Since the time of the early pioneers of neuroscience, a highly effective approach toward studying brain networks has been to focus on its cellular elements, the distinct neuronal subtypes. In the case of inhibitory interneurons within the hippocampus — a brain structure known to play key roles in certain forms of learning and memory as well as in a variety of brain disorders — research aimed at reconstructing the neuronal machine from its diverse cellular elements revealed that distinct interneurons form synapses only with certain specific parts of the post synaptic cells, such as the initial segment, cell body, proximal or distal dendrites. For example, chandelier (axo-axonic) cells release the inhibitory neurotransmitter GABA selectively onto the axon initial segments of the excitatory pyramidal cells, whereas basket cells innervate the cell body and the proximal dendritic region, and still other interneuron types are specialized to form synapses exclusively on distal dendrites. Because the different subcellular domains of neurons serve different roles, the selective innervation patterns of interneuronal types reflect are markable functional division of labor. More recently, it has been recognized that, in addition to the subcellular, spatial specificity of the individual interneuronal subtypes, there is also a great deal of temporal specificity. For example, chandelier cells, basket cells and the dendritically projecting interneurons release GABA at different times during behaviorally relevant network oscillations such as the theta rhythm or sharp wave ripples. This closely inter-twined spatiotemporal specificity of neuronal network organization is captured by the term “chronocircuitry.”

In my talk, I discussed new results regarding chronocircuit organization and control in the hippocampus. First, concerning spatial specificity, I showed new data that demonstrate that interneurons do not generate a homogenous or “blanket” inhibition as previously thought, but they form highly specific subcircuits with specific types of pyramidal cells, creating asymmetric inhibitory interactions between separate output channels from the hippocampus. Second, regarding the temporal aspects of chronocircuit organization, I presented evidence for overarching “meta”rules for subtype-specific firing, for example, in the form of frequency invariant temporal ordering of the firing of basket cells and dendritically projecting cells. Third, I showed that neuromodulatory processesal so fit into the newly recognized heterogeneity of hippocampal microcircuits. For example, we recently recognized that cannabinoid modulation of certain ion channels called HCN channels selectively occurs in a precisely defined subset of hippocampal pyramidal cells. New insights into circuit organization are increasingly made possible by sophisticated high-throughput techniques that generate cell type specific data on neuronal circuits with unprecedented detail, accompanied by rapid increases in computing power. In the fourth part of my talk, I discussed our efforts to construct strictly data driven, full-scale (1:1) computational models of the hippocampus in order to gain quantitative insights into the roles of the various constituent cell types in chronocircuit operations. These full-scale models are being used to test specific mechanistic hypothesis — for example, concerning the role of basket cells in the generation of the hippocampal theta rhythm. In addition, I also discussed the possibility to rationally derive simpler computational models that can be run on personal computers from the full-scale models. Finally, I presented recent results to show that it is now possible to control abnormal chronocircuit behavior — for example, in experimental temporal lobe epilepsy, using on-demand optogenetic interventional strategies. These new closed-loop optogenetic approaches offer the possibility to intervene in malfunctioning chronocircuits with unparalleled spatial, temporal and cell type-specificity.

Work in my laboratory is funded by the NIH, NASA and NSF.