Abstract
Neurons are among the most highly specialized cell types in the body, capable of receiving and integrating inputs across vast spatial and temporal scales to encode and transmit information. Despite this remarkable specialization, they are fundamentally biological entities equipped with the same genome as any other somatic cell type. Neuronal identity is specified and progressively refined through the sequential actions and graded expression of Transcription Factors which decode the genome to facilitate these transitions. Even once a neuron is terminally differentiated, Transcription Factors continue to serve as a nexus between the extracellular environment and intracellular milieu, enabling neurons to encode patterns of activity into transcriptional output. The roles of many fate-determining and activity-dependent Transcription Factors have been extensively characterized in the developing and mature brain, but there still remain a significant number of Transcription Factors present in the adult mouse brain with no known neuronal function. The challenge of studying novel Transcription Factors in the postnatal brain is twofold. First, many Transcription Factors lack conditional knockout lines that enable the separation of essential developmental functions of these proteins from their specific roles in neurons. Second,
many Transcription Factors are parts of larger gene families with overlapping biological functions whose members can often compensate for their loss. Therefore, the study of Transcription Factor function in vivo requires a cell type-specific manipulation that is capable of overcoming redundancy inherent in transcription factor families. In this thesis, I describe my use of the CRIPSR Interference (CRISPRi) system to simultaneously address both of these problems. Using a recently generated conditional CRISPRi knock-in mouse, I establish a streamlined approach to knocking down Transcription Factors individually or in combination in the postnatal mouse brain characterized by reliably high knockdown efficiency and penetrance. I demonstrate how this CRISPRi approach can be used to gain new insight into the relationships between paralogous Transcription Factors using the Regulatory Factor X (RFX) family as an example. Finally, I show that this approach may come with off target-effects that have not yet been described in CRISPRi experiments and offer a means to identify them.
In the final chapter, I use previously generated gene expression and chromatin accessibility data to predict a role for the Krüppel-like Family (KLF) of Transcription Factors in regulating postnatal gene expression. Using the CRIPSRi approach described in the previous chapter, I validate this prediction by demonstrating how compensatory pairs of KLF paralogs bidirectionally regulate a set of shared targets during early postnatal development. I propose a competitive binding model to explain the developmental activity of the KLF family and offer target genes through which these factors could exert their previously described opposing effects on axon outgrowth. I suggest that the transcriptional switch facilitated by the KLF family could underlie postnatal transition from axon growth to synaptic refinement. Together with my experiments characterizing the RFX family, this work demonstrates how CRISPRi can be used as a tool for in vivo studies of Transcription Factor paralogs. The data shared here, along with future studies, could shed light on how the evolutionary expansion of Transcription Factor families has contributed to the robustness of gene expression in neuronal development and the growth of gene regulatory networks in neurons.