Project Summary/Abstract: The annotation of the mammalian protein coding genome is alarmingly incomplete. Despite the massive acceleration and widespread use of transcriptomic approaches to understand biological processes, we still do not fundamentally understand RNA translation at its most basic level. Traditional definitions that categorize an RNA as either protein coding or non-coding, are currently incompatible with recent findings from genome wide translatome studies. We hypothesize that a combination of technological barriers and an adherence to dogmatic assumptions of what constitutes an open reading frame (ORF) and protein coding RNA, have severely constrained identification of a plethora of novel regulators of biological processes. We term this uncharacterized material genomic dark matter. In this proposal, we aim to systematically identify and functionally uncover its true contribution to the inflammatory response. By utilizing ribosome profiling in steady state and activated macrophages, we have identified “non-coding” RNA undergoing robust translation. Furthermore, we uncovered widespread polycistronic translation of multiple ORFs within classically annotated protein coding genes. To reveal the functional contribution of these alternative ORFs and delineate their role from that of their gene’s annotated ORF, we propose to conduct a parallel loss and gain of function inverse screening approach to identify novel proteins that contribute to the inflammatory response. Furthermore, although protein coding genes are thought to solely function by providing a message for protein synthesis, we have identified a class of mRNAs that are highly differentially expressed following bacterial stimulation but are not translated. This philosophically questions the very classification of a protein coding gene. By combining transcriptional silencing and ORF disruption studies, we aim to decouple the functional contribution of a gene’s RNA from its coding potential. In addition, although current ribosome profiling technologies are ill suited to discover unannotated transcripts undergoing translation, we have identified a plethora of mRNA derived from atypical “trans-splicing” between two different transcripts that appear to encode novel chimeric proteins. By developing an innovative technological pipeline, RiboFusionSeq, we will generate the first rigorous identification platform for coding chimeric RNAs. Furthermore, to uncover the capacity of intra-transcript circularization via “back-splicing” to encode novel protein, RiboCircSeq will be established. Using knockdown approaches specifically targeting trans- and back-splicing events, we will conduct the first functional screening of proteins derived from atypical splicing and interrogate their contribution to immunity. Finally, we will generate animal models to test the mechanistic and physiological importance of our findings in inflammation and disease. Together, these studies will 1) provide a transformative level of resolution on the protein coding genome during the immune response, 2) establish a new paradigm for the functional annotation of mammalian genomes, 3) identify a plethora of new molecules for further investigation and 4) have far reaching implications to understanding the processes underlying all human diseases.

Project Narrative: The identification and classification of the genome is an essential foundation for understanding human physiology. This work will explore the remaining uncharted corners of the genome that are activated during inflammation. By discovering the functions of these unknown regulators of immunity, we hope to uncover novel mechanisms underpinning all human disease and identify new therapeutic targets never before investigated.