Department of GeneticsHarvard Medical School
New Research Building, Room 0356
77 Ave. Louis Pasteur
Boston, MA 02115
Visit my lab page here.
Animal genomes endow cells and circuits with the ability to form life-long memories. How does a genome orchestrate this dynamic interaction of neurons with experience? The genome responds to experience by unleashing bursts of new gene expression that rewire circuits to store long-term memories. These bursts of gene expression rely on an exceedingly complicated network of neuronal activity-regulated transcription factors. It is neither known why such an extensive network is necessary nor how it works. We are taking a two-pronged approach to understand how this transcriptional network rewires neuronal circuits. First, we are applying recently developed genomics and systems biology approaches to understand how the activity-regulated transcriptional network responds to increases in neuronal firing rates. At the same time, we are establishing brain slice and in vivo experimental systems in which we can manipulate the transcriptional network in an intact circuit, using electrophysiological, optogenetic, and behavioral tools to assess how these manipulations affect neuronal circuit rewiring.
One major focus is on understanding how the activity-regulated transcriptional network, which contains at least 20 transcription factors, activates transcription when neurons within a circuit increase their firing rate. Several specific transcription factors within the network have been studied extensively, but very little is understood about how the network functions as a whole. Remaining undefined are its structure, its dynamical input-output properties, and how it differs from one neuronal subtype to the next. We are approaching these questions using primary cultures of mouse cortical neurons, where we control neuronal firing rates using the light-activated ion channel Channelrhodopsin. In this context, we manipulate the transcription factors themselves using genetic and RNAi-based strategies, and we assess genome-wide transcriptional responses using chromatin immunoprecipitation sequencing, RNA sequencing and high-throughput quantitative PCR. Finally, we interrogate activity-regulated genomic DNA elements functionally, using a high-throughput screening approach built on viral libraries of barcoded reporter genes.
Understanding the activity-regulated transcriptional network will give us an unprecedented molecular toolkit for labeling and manipulating neurons in relation to their activity. Our second major focus is on applying this toolkit to study structural plasticity in neuronal circuits and behavior in whole animals. One key step will be to use transcriptional activation to label specific neurons that are activated during memory acquisition and whose newly formed connections store a “memory trace.” This approach will be facilitated by the fact that robust long-term memories can be formed in single training sessions in behavioral paradigms such as auditory fear conditioning. Upon labeling the relevant neurons, we will genetically silence them to establish their importance for memory storage. Ultimately, we will manipulate the activity-regulated transcriptional network within these identified neurons to probe the cell biological mechanisms of memory formation using electrophysiological, imaging, and behavioral assays. By manipulating the activity-regulated network in specific neurons, it may be possible to strengthen or weaken memory formation, bias a circuit toward storing certain types of memories, or erase specific memories.
Last Update: 8/22/2013