Eukaryotic RNA processing and quality control involves factors that add 3' end modifications on RNA transcripts. However, the importance of most modifications of small non-coding RNAs (sncRNAs) is poorly understood. The phosphodiesterase Usb1 and the non-canonical poly(A) polymerase TENT2 are known to modify the 3’ ends of RNA polymerase-III (pol-III) transcripts U6 and 7SL, respectively. However, Usb1 and TENT2’s targets have not been systematically identified. To better understand the scope of 3' end modifications of sncRNAs, I analyzed differences in post-transcriptional 3' end modifications between pol-III and pol-II sncRNAs and found that adenylation can be observed at the mature 3’ end of most pol-III transcripts at a level significantly higher than for pol-II transcripts. For all pol-III transcripts, only a subset of the population is adenylated at steady state, and adenylation is generally restricted to a single adenosine. By quantifying differences in 3' end adenylation via 3' end sequencing in the presence or absence of TENT2, I identified several pol-III transcripts as targets of TENT2, including 7SL, U6atac, and Y RNAs. As Usb1 is known to add 3’ end cyclic phosphates to U6 snRNA, we hypothesized that Usb1 may be responsible for modifying additional pol-III transcripts. However, my experiments did not show evidence for cyclic phosphates on additional tested pol-III transcripts. My results demonstrate that adenylation can be observed for most pol-III sncRNAs and that TENT2 is responsible for the adenylation of a subset of these RNAs, suggesting a general role for 3’ monoadenylation in pol-III transcript processing or function. 

The RNA binding protein tristetraprolin (TTP) is an mRNA destabilizing factor that regulates the stability of transcripts involved in promoting inflammation. TTP regulates mRNA stability through recruitment of decay factors involved in deadenylation, decapping, exonucleolytic decay, and translation repression. TTP activity is regulated by post-translational modifications with nearly 30 reported sites of phosphorylation; however, the functional effects of only two sites are well characterized. Previous work revealed the conserved TTP CNOT1 Interaction Motif (CIM) is an important structural element that promotes the association with the major deadenylase complex, CCR4-NOT. In the work described in Chapter 2, I studied how the TTP CIM promotes mRNA decay cooperatively with other regions of the TTP protein. Using tethered decay assays, I find that mutation of conserved tryptophan residues, which associate with the CCR4-NOT subunit CNOT9, in addition to removing the CIM causes stabilization of target mRNA. Pull-down assays of TTP mutants reveal that CCR4-NOT recruitment is facilitated in a cooperative manner by the TTP-CIM and tryptophan residues. Furthermore, I find, contrasting previous reports, that the p38 MAPK pathway does not target a conserved serine residue within the TTP-CIM and that the CIM remains active in conditions where TTP is targeted by the downstream p38 MAPK kinase, MK2. These results suggest that efficient regulation of TTP requires multiple signaling pathways. Difficulties examining TTP under native conditions has been a limitation of previous studies. In the work described in Chapter 3, to facilitate studies of TTP in a more endogenous setting, I generated TTP-/- RAW 264.7 mouse macrophage cells using a CRISPR-Cas9 targeting approach. These were used to generate lentiviral add-back cells expressing TTP from the endogenous TTP promoter. Unfortunately, exogenous TTP expression was unable to reach the levels seen from the endogenous locus in RAW 264.7 cells suggesting that additional enhancer element(s) acting on the TTP locus are necessary to produce an efficient TTP-expression response. Additional repair template-mediated CRISPR-Cas9 gene editing techniques using selectable markers are also reported here as a promising tool to generate endogenous TTP mutants in RAW 264.7 macrophage cells.

The zinc finger protein Tristetraprolin (TTP) promotes translation repression and degradation of mRNAs containing AU-rich elements (AREs). While much attention has been directed towards understanding the decay process and the machinery involved, the translation repression role of TTP has remained poorly understood. My thesis research identified the cap-binding translation repression 4EHP-GYF2 complex as a co-factor of TTP. Immunoprecipitation and in vitro pulldown assays demonstrate that TTP associates with the 4EHP-GYF2 complex via direct interaction with GYF2, and mutational analyses show that this interaction occurs via conserved tetraproline motifs of TTP. Mutant TTP with diminished 4EHP-GYF2 binding is impaired in its ability to repress a luciferase reporter ARE-mRNA. 4ehp knockout mouse embryonic fibroblasts (MEFs) display increased induction and slower turnover of TTP target mRNAs as compared to wild-type MEFs. This work highlights the function of the conserved tetraproline motifs of TTP and identifies 4EHP-GYF2 as a co-factor in translational repression and mRNA decay by TTP.

The human genome encodes for two TTP homologs, BRF1 and BRF2, both of which are capable of ARE binding and decay activation. I observed that TTP family proteins are differentially regulated during serum-activated G0-G1-S transition in NIH 3T3 cells. The G0-G1-S cell cycle transition is regulated at multiple levels to avoid erroneous or mistimed cell proliferation. Transition over the G0-G1-S cell cycle requires repression of Retinoblastoma (RB) proteins, which are central components of the CDK-RB-E2F checkpoint pathway controlling in this transition. I identified ARE-containing Rb, Rbl1 and Rbl2 mRNAs encoding retinoblastoma proteins as targets of degradation by TTP family proteins. Depletion of TTP family proteins results in reduced expression of E2F transcription targets, and slower transition out of quiescence compared to control conditions. Together, this work demonstrates the involvement of mRNA decay in cell cycle progression.

The non-coding 7SL RNA transcript complexes with six SRP proteins to form the Signal Recognition Particle. 7SL RNAs are produced with encoded U-tails of variable length due to their transcription by RNA polymerase III, and about 80% of endogenous 7SL RNAs receive one or two adenines at their 3’ ends by noncanonical poly(A) polymerases including TENT2. However, the relationship between SRP protein binding, 7SL RNA stability, and 3’ end tailing dynamics is not understood. To begin to investigate this, I established an exogenous 7SL1 RNA expression system with point mutations serving as barcodes for distinguishing from endogenous 7SL RNAs. Two independent exogenous barcoded 7SL1 RNAs, however, showed less 3' end adenylation and more uridylation than endogenous 7SL RNAs. A subsequent titration experiment revealed increased 3’ end adenylation and decreased uridylation at lower expression levels but still did not fully match endogenous 7SL RNAs even when expressed at below 10% of endogenous levels. TENT2 knockout caused almost complete loss of adenylation of the exogenous barcoded 7SL1 RNAs but no change in accumulation, suggesting that adenylation is not important for stability. Mutations impairing association of 7SL1 RNA with SRP54 showed no effects on accumulation or uridylation but moderate increases in adenylation, suggesting SRP54 binding does not impact 7SL RNA 3’ end processing or stability in a major way. Mutations known to impair association of 7SL1 RNA with SRP9, however, demonstrated increased uridylation and decreased adenylation and accumulation, suggesting SRP9 assembly impacts 7SL RNA 3' end tailing and stability.

Chapter I is an introduction to the unifying theme of my thesis, competition between translation initiation and mRNA decay with a focus on my proteins of interest.

Chapter II is a reprint of a review co-authored by myself and Donghui Wu. The review covers structural insights into the mechanisms of decapping, catalytic decapping enhancers, and enhancers that function primarily by repressing translation.

Chapter III focuses on the role of the decay factor Lsm4 in regulating Processing Bodies (PBs). PBs are cytoplasmic foci that contain mRNA decay factors decay intermediates. PBs are lost when human cells are depleted of Lsm4, which contains an arginine/glycine rich RGG C-terminal domain that has been shown to undergo arginine dimethylation. Using immunofluorescence and Lsm4 mutants, I show that arginine methylation, likely by the methyltransferase PRMT5, in the RGG domain of Lsm4 is critical for PB accumulation. Assays for mRNA decay and translational activity, and mass spectrometry analysis all suggest that the RGG domain is not required for efficient decay, translation inhibition or decay factor recruitment.

Chapter IV shows the results of a collaboration investigating the decapping enhancer PNRC2. PNRC2 can interact with and stimulate the activity of the Dcp1/Dcp2 decapping complex. Disrupting this interaction leads to a loss in decapping stimulation. Chapter V focuses on PABP, a highly conserved protein that binds to the 3’ poly(A) tail of nearly all mRNA and stimulates translation. We wondered if modifications to PABP could regulate mRNP function similar to how histone modifications regulate transcription. I used PABP mutants and assays designed to identify protein modifications to investigate this question.

Chapter VI is a look at the most pressing questions arising from my research.

All RNAs in the cell are monitored by robust molecular machinery that ensures its fidelity and quality. Maintenance of the integrity of the transcriptome is vital to the proper functioning of the cell. Dysfunctions in the machinery responsible for this quality control can lead to physiological problems for the cell and wide-ranging diseases in the context of a whole organism. A deeper understanding of these systems is critically important to illuminate the workings of the cell, multi-cellular organisms, and to provide new therapeutic handles to treat human disease. The cell contains multiple overlapping quality control pathways to monitor RNAs that are tailored to the type and purpose of a particular RNA. RNAs that do not code for proteins, non-coding RNAs (ncRNAs), are increasingly being understood for the dynamic trimming and tailing that occurs in their 3’ end to control their maturation and quality. Advances in next generation sequencing have allowed researchers to examine the changing 3’ ends of ncRNAs at nucleotide resolution, but robust, open source, and publicly available tools to analyze this data have been missing from the field. In chapter 2 of this dissertation, I outline a bioinformatic tool dubbed “Tailer” that allows for visualization and quantification of this kind of data. This tool is thoroughly validated on public datasets and faithfully recapitulates the findings of the original studies from which they were drawn. Another key hub for RNA quality control, specifically for mRNAs, is translation. When translating a dysfunctional mRNA or an mRNA in a difficult context, a ribosome can run into regions that cause a local ribosome stall. This allows a trailing ribosome to collide, signaling to the cell that the nascent protein and mRNA template need to be degraded. In chapter 3, I present evidence that Angel1, a 2’,3’ cyclic phosphatase, is a novel rate-limiting factor for this ribosome-associated quality control (RQC) pathway. Angel1 associates with proteins important for this process and nucleotide sequences that have been implicated in ribosome stalling. Angel1 depletion also stabilizes reporter substrates that are targets of this pathway. I also show evidence that N4BP2 is the human ortholog of the RQC endonuclease identified in Saccharomyces cerevisiae and Caenorhabditis elegans. Lastly, in chapter 4, I examine Angel2, a protein homologous to Angel1, that was recently identified as a 2’,3’ cyclic phosphatase. Using discovery-based techniques, I look for clues that could suggest Angel2’s function. With IP-MS/MS I find that Angel2 associates with components of the tRNA splicing ligase. Using eCLIP, I find evidence that Angel2 directly associates with tRNAs but is not necessarily enriched for the subset of tRNAs that undergo splicing. Finally, with RNA-seq, I find that Angel2 depletion upregulates mRNAs with a low fraction of rare codons and codons decoded by intron-containing tRNAs. These studies provide new insights into RNA quality control and provide a leaping off point for future studies into the modification and processing of RNA 3' ends.

During times of unpredictable stress, organisms must adapt their gene expression to maximize survival. Along with changes in transcription, one conserved means of gene regulation during conditions that quickly represses translation is the formation of cytoplasmic phase-separated mRNP (messenger ribonucleoprotein) granules. Two well-known stress-induced mRNP granules are processing bodies (P-bodies) and stress granules. Previously, we identified that distinct steps in gene expression can be coupled during glucose starvation stress as promoter sequences in the nucleus are able to direct the subcellular localization to or excluded from the phase-separated granules and translatability of mRNAs in the cytosol. During my Ph.D., I have been investigating the underlying mechanisms of how promoter dictates the cytoplasmic fate of mRNAs during glucose starvation. Chapter 1 is a background introduction of the questions I asked. In chapter 2, I led a discovery of a promoter-directed mechanism mediated by AAA+ proteins Rvb1/Rvb2 that couple the transcription, mRNA cytoplasmic localization, and translation of select genes during glucose starvation. In chapter 3, colleagues and I reviewed recent findings regarding stress-related phase-separated granules and shared our insights. Chapter 4 describes 3 projects that aimed to answer the promoter-dictation question from other perspectives, such as expression timing and promoter motifs. Finally, Chapter 5 discusses insights and future directions for my thesis work. I anticipate that these studies will provide us with more insights into complex gene regulation during stressful and pathological conditions.