RNA-binding proteins (RBP) play essential roles in shaping gene expression programs by controlling the maturation, modification, transport, stability, and degradation of all classes of RNAs, and by mediating the functions of non-coding RNA molecules. The ubiquitous nature of RBP-mediated gene regulation means these proteins have the potential of impacting virtually every biological process. Perhaps not surprisingly, mutations in the genes that encode them are being increasingly associated with human diseases. Nevertheless, the mechanisms through which RBP-centered pathways regulate gene expression or the impact they have on cellular behavior remains incompletely understood. A number of groups have started to tackle this problem by employing RNA-pull-down methods as a way of identifying the full repertoire of RBP in living cells. In addition, the development and application of next-generation sequencing-based methods such as CLIP, 3' end-seq, m6A-seq, among others have started to provide insights on the target specificity and mode of action of individual proteins in this catalog. These studies have highlighted an unanticipated complexity in RBP-centered gene regulatory networks and the need for approaches to experimentally determine their role in a systematic manner. CRISPR-based high-throughput screens offer a unique opportunity to address this need by providing a fast and effective way of simultaneously querying the function of multiple genes. Nonetheless, Cas9 off-target activity remains a concern and can contribute to significant noise in the data, leading to the occurrence of a large number of false-positives and false-negatives, and undermining the ability of gRNA libraries to assign function to genes that have subtle but physiologically relevant roles in cells. Over the past 5 years we have gathered extensive expertise in the use of CRISPR/Cas9 technology to dissect gene function in vitro and in vivo. We have also developed computational and experimental tools for library design and construction in the context of CRISPR large-scale screens. Here, we propose use these expertise to i) develop and test new gRNA design rules that maximize targeting specificity and library sensitivity; and ii) use these rules to generate novel RBP-wide CRISPR libraries that can be used to experimentally dissect RBP networks in a variety of cellular contexts.
Aim 1 : Novel gRNA design rules for sensitive high-throughput CRISPR screens. Our own published work has shown that widely used computational approaches to design gRNAs consistently fail to identify off-target sites with perfect or near perfect complementarity to the guide. We also demonstrated that as a consequence, these promiscuous gRNAs can be given misleadingly high specificity scores and that their experimental use can lead to the generation of complex genomic rearrangements. We have now gathered preliminary data suggesting that promiscuous gRNAs-often containing multiple perfect off-target sites in the genome-are ubiquitous in published CRISPR libraries and significantly contribute to the occurrence of false-negative and false-positive hits in high-throughput experiments. To overcome this problem we will use the computational approaches we have previously published to fully characterize the impact of off-target activity on the accuracy and sensitivity of CRISPR screens and use this data to develop novel gRNA designs. We will test these against those of previously published libraries. Successful completion of this aim will lead to the generation of more effective experimental and computational genome-editing tools for high-throughput screening, which can be used for the generation of genome-wide/focused libraries and applied to research areas outside our own.
Aim 2 : RBP focused CRISPR libraries to dissect RNA pathways in mammals. To begin to systematically characterize the roles of RNA-centered pathways in mammals, we will use the design rules defined in the previous aim to generate focused CRISPR libraries against mouse RBPs. We will employ these libraries in screens in embryonic stem carrying an Oct4-GFP reporter to define how individual proteins in this gene set contribute to maintenance of cell identity. Genes whose disruption leads to significant loss of GFP, a surrogate marker for pluripotency, will be selected for further biochemical characterization to understand how they impact gene expression. Together, the results gathered in this aim will produce the first systematic view of the impact of RNA pathways on the behavior of highly specialized cells. These results represent a first step to understand how RBPs can impact gene expression and ultimately the ability of stem cells to self-renew and maintain cell identity in the absence of differentiation factors. Uncovering these regulatory mechanisms has critical implications not only for human development and tissue homeostasis, but also to other diseases where self-renewal becomes independent of external cues. Although we propose to do our initial studies in the context of mouse stem cells, we believe the libraries we propose to generate here will represent invaluable discovery tools to dissect the role of RNA pathways in the context of basic, biomedical, and clinical research at large.
