Mechanism of Translation Regulation of BTG1 by eIF3 Master’s Thesis Presented to The Faculty of the Graduate School of Arts and Sciences Brandeis University Department of Biology Amy S.Y. Lee, Advisor In Partial Fulfillment of the Requirements for the Degree Master of Science in Biology by Shih-Ming (Annie) Huang May 2019 Copyright by Shih-Ming (Annie) Huang © 2019 ACKNOWLEDGEMENT I would like to express my deepest gratitude to Dr. Amy S.Y. Lee for her continuous patience, support, encouragement, and guidance throughout this journey. I am very thankful for all the members of the Lee Lab for providing me with this caring and warm environment to complete my work. I would also like to thank Dr. James NuñeZ for collaborating with us on this project and helping us in any shape or form. iii ABSTRACT Mechanism of Translation Regulation of BTG1 by eIF3 A thesis presented to the Department of Biology Graduate School of Arts and Sciences Brandeis University Waltham, Massachusetts By Shih-Ming (Annie) Huang REDACTED iv TABLE OF CONTENTS REDACTED v LIST OF FIGURES REDACTED vi INTRODUCTION I. Gene Regulation All cells in our bodies contain the same genome, but distinct cell types express very different sets of genes. The sets of gene expressed under specific conditions determine what the cell can do, by controlling the proteins and functional RNAs the cell contains. The process of controlling which genes are expressed is known as gene regulation. Any step along the gene expression pathway can be controlled, from DNA transcription, translation of mRNAs into proteins, to post-translational modifications. Compared to transcription regulation, translation regulation can control existing mRNAs, allowing more rapid changes in the cellular concentration of proteins. Thus, translational regulation plays a crucial role in maintaining homeostasis and controlling cell growth and proliferation (Hershey et al. 2012). II. Translation Regulation in Eukaryotes Translation can be divided into initiation, elongation, termination, and ribosome recycling. The canonical mechanism of translation initiation includes the formation of 43S preinitiation complexes, the attachment of 43S complexes to mRNA, and the ribosome scanning of mRNA 5′ untranslated regions (UTRs) (Jackson et al., 2010). The 43S preinitiation complex is formed when the methionine-tRNA is delivered by eIF2 to the P site of the 40S ribosomal subunit (Sonenberg and Hinnebusch, 2009). The complex is then recruited to the 5′ end of the mRNA by eIF3 and eIF4 (Kumar et al., 2016). After attachment, the 43S preinitiation complex scans the 5′ UTR and recogniZes the AUG start codon. Through scanning, the secondary structure of the 5′ UTR is unwound while the ribosome moves along it (LópeZ-Lastra et al., 2005). Once the initiation codon is recogniZed, the 60S subunit joins to form the 80S initiation complex for subsequent elongation. 1 Notably, the interactions between the eukaryotic initiation factors, ribosomes, and mRNA for the formation of the initiation complexes are potential steps for regulation. Regulation by RNA-binding Proteins RNA-binding proteins (RBPs) are proteins that can bind to single stranded or structured RNAs. RBPs can direct mRNA-specific translation regulation through interactions with the 5′ or 3′ UTRs. For example, the gamma interferon-activated inhibitor of translation (GAIT) complex binds to specific 3′ UTR elements and suppresses translation by inhibiting the recruitment of the 43S translation initiation complex (Mukhopadhyay et al., 2009). Another RBP, CNBP/ZNF9 binds to G-rich elements in target mRNAs to prevent G-quadruplex formation. Thus, promoting translation by stabiliZing mRNA structures (Benhalevy et al., 2017). With its versatile modes of action, RBPs can regulate many aspects of mRNA stability, localiZation, and export. Regulation by Secondary Structure Other than serving as a translation regulatory element by interacting with RBPs, secondary structure in the 5ʹ UTR of an mRNA can also directly play a functional role in translation regulation. For example, genes encoding protooncogenes, growth factors, and differentiation often have stable secondary structures in the 5ʹ UTR and are poorly translated under normal conditions (Davuluri et al., 2000). It is thought that since efficient translation requires scanning of the 5ʹ UTR by the preinitiation complex, stable secondary structures inhibit translation by blocking ribosome movement (Rozen et al., 1990; Pickering and Willis, 2005). Upstream Open Reading Frames Regulate Gene-Specific Translation Upstream open reading frames (uORF) are one of the most common regulatory RNA elements. In fact, 49% of the human transcriptome contains uORFs (Suzuki et al., 2000). uORFs are short ORFs within the 5' UTR, upstream of the main coding sequence (CDS). Typically, uORFs 2 are thought to be inhibitory to translation of the downstream CDS. When an uORF is recogniZed by a scanning ribosome, the translation machinery can start translation at this inappropriate start codon, and then dissociates at the next stop codon. After this, the ribosome can either stall during elongation to block recruitment of additional ribosomes, trigger mRNA decay, or can continue to scan to reinitiate downstream translation at the appropriate start codon (Meijer and Thomas, 2002). All these different mechanisms suggest that uORFs can be major regulators of translation. III. Eukaryotic Initiation Factor 3 (eIF3) eIF3 is a multiprotein complex with 13 subunits (eIF3a to eIF3m) (Figure 1). It is involved in almost every step of translation initiation. During canonical translation, eIF3 interacts with many of the eukaryotic initiation factors for ribosome recruitment, attachment of 43S complexes to mRNA, and the scanning of the 5′ UTR (Yin et al., 2018). Besides its prominent role in canonical translation, the subunits of eIF3 have been shown to have specialiZed functions. For example, eIF3a plays a crucial role in regulating cell cycle progression and cell proliferation (Dong et al., 2009). Its expression also has been linked to different types of cancer, although the exact mechanism by which eIF3 controls cell proliferation remains unknown (Pincheira et al., 2001; Chen and Burger, 2004). Furthermore, downregulation of eIF3b is linked to cell arrest and apoptosis (Liang et al., 2012; Choi et al., 2017). Elevated expression of eIF3b has been detected in several tumors (Wang et al., 2013; Choi et al., 2017). eIF3d has also been shown to play an oncogenic role in many different types of cancer (Fan and Guo, 2015; Lin et al., 2015). Recent research has shown that eIF3d can serve as a cap-binding protein for eIF4E-independent translation by binding to the 5′ cap structure of specific mRNAs (Lee et al., 2016). Additionally, eIF3g contains an RNA recognition motif that is responsible for mediating the interaction of the 3 43S preinitiation complex and the mRNA (Zhou et al., 2008). eIF3g has also been shown to interact with nuclear and cytoskeleton proteins (Zheng et al., 2016). Figure 1. EM structure of mammalian eIF3 Ribbon representation of the eIF3 core subunits in the context of the 43S preinitiation complex (Des-Georges et al., 2015). IV. Translation Regulation by eIF3 Previous research by Lee et al. showed that eIF3 is responsible for regulating cell proliferation and differentiation by interacting with select mRNAs, including c-JUN and BTG1(Figure 2). As shown by photoactivatable ribonucleoside-enhanced crosslinking and immunoprecipitation (PAR-CLIP), eIF3a, b, d and g subunits bind to the 5′ UTR of specific mRNAs (Figure 3). By using selective 2′-hydroxyl acylation analysed by primer extension (SHAPE), Lee et al. experimentally determined the secondary structure of eIF3-binding sites and showed that eIF3 binds to RNA stem–loop elements in the 5′ UTR of the target mRNAs. However, native agarose gel electrophoresis shows that eIF3 interacts with c-JUN and BTG1 with different modes of action. eIF3 interacts with c-JUN both in vivo and in vitro, indicating this is a binary- 4 interaction. In contrast, eIF3 only interacts with BTG1 in vivo but not in vitro, suggesting that another co-factor(s) is needed for the interaction to occur. Figure 2. Validation of eIF3 PAR-CLIP targets eIF3 PAR-CLIP targets validated by eIF3 immunoprecipitation followed by RT-PCR (Data from Lee et al., 2015). Figure 3. eIF3 subunits crosslinked to RNA SDS gel showing eIF3 subunits crosslinked to RNA from PAR-CLIP (Data from Lee et al., 2015). 5 To investigate the effect of the eIF3–RNA interaction on translation, a luciferase reporter assay was used to measure translation driven by the 5′ UTR of BTG1 or c-JUN with and without eIF3-binding sites. Data showed that eIF3 serves as an activator for c-JUN translation and a repressor for BTG1 translation (Figure 4). Together, Lee et al showed that besides serving as a scaffold protein during translation initiation, eIF3 also acts as transcript-specific translation regulator by binding to specific mRNAs with different modes of action. Figure 4. Luciferase rePorter assay showing eIF3 as a translation regulator Luciferase activities measuring translation efficiency of mRNA driven by the c-JUN 5′ UTR or BTG1 5′ UTR with or without eIF3-binding site (Data from Lee et al., 2015). V. BTG anti-Proliferation factor 1 (BTG1) BTG anti-proliferation factor 1(BTG1) is an antiproliferative gene that is involved in regulating cell growth and differentiation. It is part of the BTG/Tob family, which share a conserved N‐terminal BTG domain that has been implicated in protein-protein interactions (Matsuda et al., 2001). BTG Anti-Proliferation Factor 2 (BTG2), a homolog of BTG1, has been shown to control cell cycle progression by responding to DNA damage (Rouault et al., 1996). Despite their differences in their role in cell cycle regulation, BTG1 and BTG2 share many similarities in terms of their structure and expression regulation. 6 The BTG1 gene in human is located on chromosomes 12q22 while BTG2 is located on 1q32 (Rimokh et al., 1991; Rouault et al., 1996).