UC San Diego

The mammalian cerebral cortex is composed of a diversity of neuronal cell types with distinct morphology, molecular composition, and electrophysiological properties. These neurons connect to one another to form complex microcircuits that underlie brain cortical processing. Thus, deciphering the precise input and output connectivity patterns of different neuronal cell types is conducive to understanding their functional roles in cortical processing. To this end monosynaptic rabies tracing has been widely used for cortical circuit tracing studies and has had great impact on the understanding of neural circuit organization. Still, the advent of single-cell genomic technologies has unveiled that the extent of neuronal diversity may be much greater than originally imagined, raising new questions about the connectivity patterns of these more precisely defined cell subtypes. Our understanding of cortical circuit organization could benefit from higher throughput methods of assigning inputs to neuronal cell types and the ability to assign cells to finer subtypes. Chapter 1 explores the feasibility of combining monosynaptic rabies tracing with single-nuclei RNA-sequencing (snRNAseq) to identify the transcriptomic cell types that provide presynaptic inputs to defined populations of neurons. We found that, despite global and cell-type-specific rabies-induced transcription changes, rabies-infected cortical cells can still be classified according to established transcriptomic cell types when utilizing transcriptome-wide RNA profiles. In Chapter 2, we characterize the interlaminar synaptic connectivity of mouse primary visual cortex (V1) at the transcriptomic level using the newly developed method Single Transcriptome Assisted Rabies Tracing (START). We found that START generates results consistent with established circuit models validating the utility of START as a circuit tracing tool. More importantly, with the improved cell type granularity achieved with transcriptomic characterization of inputs, we were able to uncover subtypes of somatostatin and parvalbumin interneurons that provide input to L2/3 and L6 CT excitatory neurons. Finally, Chapter 3 describes the efficiency of rabies transsynaptic spread from starter cells to input neurons. We found that about 40% of inputs are labeled transsynaptically. Altogether, this dissertation reveals how transcriptomically defined cell-types are organized in V1 and introduces a novel circuit tracing technique that will expand the repertoire of tools available to neuroscientists.