Poster abstracts
Poster number 158 submitted by Christina Akirtava
Unraveling the roles of 5’ transcript leaders in gene regulation
Christina Akirtava (Carnegie Mellon University, Biological Sciences), Hunter Kready (Carnegie Mellon University, Biological Sciences), Lauren Nazzaro (Carnegie Mellon University, Biological Sciences), Matt Agar-Johnson (Carnegie Mellon University, Biological Sciences), Gemma E May (Carnegie Mellon University, Biological Sciences), C. Joel McManus (Carnegie Mellon University, Biological Sciences)
Abstract:
Eukaryotic translation initiation typically follows a cap-dependent directional scanning model, and thus is limited by the 5’ transcript leader (TL) and start codon. The 5’ TL contains a variety of cis -regulatory sequences, structures and trans -acting factors which influence translation initiation efficiency (Kozak, 1984; Rojas-duran & Gilbert, 2012). Previous studies using in vivo reporters of fixed-length synthetic TLs identified upstream AUGs as major repressors of gene expression (Cuperus et al, 2017; Dvir et al, 2013; Sample et al, 2018). However, the relative contributions of these features to translation regulation in natural TLs have not been directly determined. To address this, we used two massively parallel reporter systems (MPRAs) (Noderer et al, 2014) to quantify in vivo regulation from 86% of natural yeast TLs and systematically identify the relative impacts of their sequence features on initiation in a systematic manner. First, we compared expression from genes containing alternative start sites and found huge differences in initiation regulation. We also, experimentally determined the relative strengths of all yeast Kozak contexts and used these values to develop a leaky scanning model that predicts initiation efficiency. Next, we use computational modeling to explain ~64% of translation initiation, detect key cis-acting sequence features, and quantify their effects in vivo. Additionally, we tested the TLs in an eIF2a phosphomimetic mutant that limits translation initiation. Most TLs exhibit a decrease in translation efficiency; however, a subset of TLs increase YFP expression with changes up to 4-fold. Machine learning models suggest preinitiation complexes scan the TL less effectively and are highly impacted by TL length and structure in the eIF2a mutant. Finally, we compared our results to sequence features found in a recent massively parallel in vitro study of translation initiation (Niederer et al, 2022). Together, our results identify the range and relative influence of cis-acting sequences and structures on translation initiation from native yeast TLs in vivo.
Keywords: translation, modeling, UTR