The nucleosome remodeling factor (NURF) is an evolutionary conserved ATP-dependent chromatin remodeling factor. It was first isolated from Drosophila as a complex with enzymatic activity that once recruited to nucleosome, it slides the nucleosome to provide accessibility for transcription factors. Since then, numerous works from animal models and cell lines show the role of NURF as a regulator of gene expression. NURF interacts with H3K4me3 and sequence specific transcription factors that recruit the complex to promoter regions. Whether this is the only mechanism by which NURF regulates gene expression is not known. However, other ATP-dependent chromatin remodeling complexes are known to regulate gene expression independent from transcription initiation. In order to explore the role of NURF in regulating gene expression, we utilized two genome wide approaches to map NURF binding and NURF dependent changes in chromatin structure using ChIP-Seq and FAIRE-Seq, respectively. From these analyses, we discovered that NURF broadly localizes in the genome with preferences to gene bodies and 3’ends of genes. Also, we found that NURF maintains open chromatin regions at upstream, intron and downstream of genes. These novel findings shed light on new roles for NURF complex within genes, in addition to its classical role at promoter regions. Furthermore, we discovered the function of a previously uncharacterized domain in the NURF specific subunit BPTF. We show that the N-terminal the plant homeodomain (PHD) of BPTF directly interacts with THOC4, a protein associated with RNA-pol 2. Also, we show using ChIP analyses that this interaction recruits BPTF to gene bodies. Next, we investigated functional consequences for NURF recruitment to gene bodies using Cyclin D1 (Ccnd1) gene as a model. These analyses revealed that NURF is required for normal mRNA processing and loss of NURF induces intron retention, which results in unstable transcripts. Finally, we show that the defect in mRNA processing is not specific to the Ccnd1 gene, as we observe similar defects in four other BPTF dependent genes. Together, our work uncovered new role of mammalian NURF complex in regulating gene expression through mRNA processing.

1 ABSTRACT There are many structures, suborganelles and bodies in the eukaryotic cell nucleus. These domains provide the nucleus with many specific functions. Nucleolus is specialized compartment serves to ribosomes assembly, nuclear speckles or Splicing Factors Compartment play an important role in RNA processing and best studied of them, Cajal bodies (CBs), are involved in snRNP maturation. However, non-membrane substructures are not unique for cell nucleus; processing bodies (P bodies) found in the cytoplasm are proposed to be important places in mRNA degradation pathway. This work is a compilation of four projects focused on non-membrane cellular bodies; namely, nuclear CBs and cytoplasmic P bodies. Both CBs and P bodies are dynamic structures that continuously exchange their components with surrounding environment. In addition to a widely accepted role of CBs in snRNP biogenesis, we show that the CB serves as a place where snRNPs are regenerated after each round of splicing. Thus, CBs are important nuclear compartment involved in snRNP recycling. To further characterize tri-snRNP assembly in CBs we applied kinetic experiments combined with mathematical modeling and created a kinetic model of tri- snRNP formation in the CB that determined kinetic parameters of tri-snRNP formation. Moreover, our kinetic...

Eukaryotic genes contain non-coding sequences - introns that are removed during pre-mRNA splicing by the spliceosome. The spliceosome is composed of five snRNPs (U1, U2, U4/U6 and U5) which assemble on pre-mRNA in a step-wise manner and together with additional non-snRNP proteins catalyse splicing. Mutations in splicing factors can cause severe diseases, for example a point missense mutation (called AD29) in hPrp31 (U4/U6 snRNP specific protein) induces retinitis pigmentosa, disease often leading to complete blindness. In this PhD thesis we show that the hPrp31 AD29 mutant is unstable and is not properly incorporated into spliceosomal snRNPs. In addition, the expression of the mutant protein reduces cell proliferation, which indicates that it interferes with cellular metabolism (likely splicing) and could explain the induction of retinitis pigmentosa. Next, we focus on a role of nuclear environment in pre-mRNA splicing. It was shown that new U4/U6·U5 snRNPs are preferentially assembled in non-membrane nuclear structure - Cajal body. Here we expand this finding and provide evidence that Cajal bodies are also important for U4/U6·U5 snRNP recycling after splicing. In addition, we analyzed a role of chromatin and particularly histone acetylation modulates in splicing regulation. Using inhibitor of...

One third of the world's population is infected with Mycobacterium tuberculosis, the bacterium that causes TB. During an infection, bacteria often survive host immune system attacks, which include oxidative stress conditions for bacteria growing inside macrophages. This makes treatment difficult and time-consuming. We hypothesize bacteria can adapt to environmental conditions by changing their mRNA maturation and degradation profiles. Using a model system, Mycobacteruim smegmatis, we focus on how mRNA expression is affected by oxidative stress. After construction and sequencing of RNA expression libraries, preliminary analysis showed that after three hours of H2O2 exposure most upregulated genes were related to DNA repair, while downregulated genes included transport proteins. After six hours of exposure, upregulated genes were similar to three hours and downregulated genes included tRNAs. 5' end mapping libraries were also constructed to access differential cleavage site abundance under oxidative stress conditions. We also investigated the roles RNase J may have in stress response and mRNA processing in Mycobacteria. RNase J and RNase E are thought to be the major RNases in bacteria. While most bacteria only have one of them, mycobacteria encode both in their genome, with RNase J being non-essential. We constructed a set of 4 strains (WT, RNase J overexpression, RNase J deletion, and complemented RNase J deletion) and tested their drug resistance and stress tolerance. Results suggests that RNase J deletion and overexpression alter drug sensitivity. Stress tolerance assays showed that WT is more tolerant to oxidative stress, followed by RNase J deletion strain and overexpression and complemented RNase J deletion strains, with the last two showing no growth when cultured with H2O2. Analysis of the expression profile of these strains was performed to help understand if gene expression differences are responsible for the phenotypes observed. For the complemented RNase J deletion, one operon had almost all its genes upregulated. This operon encodes a hydrogenase (Hyd3), suggesting that redox balance in the strain is perturbed.

In the fruit fly, Drosophila melanogaster, accumulation of osk mRNA at the posterior pole of the oocyte and local translation initiate assembly of the pole plasm, which is required for germ cell formation and posterior patterning of the embryo. I have used fluorescence in situ hybridization (FISH) in combination with immunofluorescence and laser scanning confocal microscopy to examine the spatial and temporal control of osk transcript localization and translation. Drosophila oocytes develop within cysts of 16 interconnected cells. One cell in each cyst differentiates to form the oocyte while the remaining cells form nurse cells that produce RNAs and proteins that are transported to the oocyte. osk mRNA is produced by the nurse cells and accumulates in the oocyte throughout oogenesis, but is only specifically localized to the posterior pole and translated during mid to late oogenesis. My studies help define distinct steps in the osk mRNA localization process. An early step in posterior localization is removal of osk mRNA from most of the cortex, leading to accumulation in the oocyte interior. This process requires microtubules, the microtubule motor protein Kinesin I, the actin binding protein Tropomyosin, and the RNA binding protein Staufen. Transcript then moves from the oocyte interior to the posterior pole through a microtubule independent process. The genes cappuccino, chickadee, spire, armitage, maelstrom, par-1 and gurken are all required for this next step in osk mRNA localization. The final capturing or tethering osk mRNA at the cortex requires an intact actin filament system, but additional components of this anchoring system remain to be identified. I also find that osk mRNA first begins to accumulate at the posterior pole during oogenesis stage 8, but protein is not detectable until stage 9. In addition, grk and par-1 mutations that block osk mRNA localization to the posterior pole and lead to transcript accumulation in the interior do not prevent translation; again, Osk protein production is only observed during stage 9 and later. These observations indicate that posterior localization is neither sufficient nor necessary to trigger osk mRNA translation, which appears to be under tight temporal control.

Sepsis is defined as potentially fatal systemic inflammation, caused by an infection. It is the leading cause of ICU mortality and the 10th leading cause of death in the United States. Several models exist to mimic this disorder, and have demonstrated differential mortality rates between the models as well as the individual animals. Previous studies have shown that elevated levels of plasma mitochondrial DNA (mtDNA) correlated with mortality in septic patients, and cell-free mitochondrial DNA can elicit toll-like receptor mediated immune responses similar to LPS-mediated septicemia. However, the role of mtDNA in the pathophysiology sepsis is still unknown. The focus of this study was to create sepsis in a mouse model using the murine Cecal Ligation and Puncture (CLP) model, and measure plasma mtDNA levels. After CLP was performed on experimental mice, blood plasma was collected 24 hours later. Elevated amounts of circulating mtDNA were detectable in the plasma using real time PCR and cytochrome B2 as a marker of mitochondria. These data were correlated with plasma IL-6 levels, which were used to predict mortality within 5 days of CLP to stratify mice into two populations of those predicted to live or die following the procedure. We also aimed to investigate the effect of mtDNA and mitochondrial debris on naïve mouse macrophages in an in vitro study of the regulation of inflammatory cytokines interleukin-6 (IL-6), tumor necrosis factor alpha (TNF-α), and interleukin-1 beta (IL-1β). In order to observe the effects of mtDNA on murine macrophages, mitochondria was purified from mouse liver and used to stimulate these cells alongside positive control, LPS. Stimulation with mtDNA and mitochondrial debris resulted in increased levels of TNF-α mRNA in lysed cells as well as their surrounding media as compared to control cells, as well as increased transcript half life as measured over four hours post stimulation with transcription inhibitor actinomycin D. The increases in mRNA half-life elicited by mtDNA were comparable to those observed after LPS addition. Stimulation also caused increased binding of TNF-α mRNA to the RNA binding protein, AUF1, as measured by immunoprecipitation of RNA-protein complexes and assayed for TNF-α binding by PCR. These results demonstrate that mitochondrial damage-associated molecular patterns regulate TNF-α mRNA expression at the post-transcriptional level through AUF1, an mRNA destabilizing factor. This is a novel mechanism that likely contributes to sepsis pathophysiology, and demonstrates the involvement of the mitochondrial fission and fusion balance and its regulation in the sepsis innate immune response.

The level to which genes are expressed depends on the rate at which the mRNA is generated, and the rate at which it is utilised and destroyed. Almost all eukaryotic mRNAs contain a stretch of adenosine nucleotides known as the poly(A) tail. The removal of the polyA tail from an mRNA, a process called deadenylation, is an important mechanism of gene expression regulation. It is the first step in the decay of the transcript, and is also linked to repression of translation. Deadenylation is predominantly catalysed by a conserved multi-protein complex called Ccr4-Not. While the poly(A) tail is a feature of almost all mRNAs, cells control the rate at which each undergoes decay by the precise targeting of Ccr4-Not in both a gene-dependent and a context-dependent fashion. Substrate-selective deadenylation is therefore a central biochemical process to the control of gene expression. It plays a pivotal role in most cellular processes including differentiation, cell cycle control and adaptation to environmental change. The inflammatory response and embryogenesis are two systems in which deadenylation has been well studied. The subject of this dissertation is the biochemical mechanisms by which mRNAs are selected for deadenylation by Ccr4-Not. Despite its importance, intact Ccr4-Not has not previously been obtained in sufficient quantity and purity for rigorous biochemical and structural analysis. Here I present the purification of recombinant Ccr4-Not. An experimental system was devised to quantify the rate and pattern of the deadenylation reaction that it catalyses in vitro. Two models of Ccr4-Not regulation were characterised in detail: the recruitment of Ccr4-Not by RNA-binding adaptor proteins, and the effect of the protein Pab1, which binds to the poly(A) tail. These have yielded insight into the features of the proteins and RNA sequences that are critical to deadenylation. In addition, a structural study of the Ccr4-Not complex was performed using electron cryomicroscopy and single-particle analysis.

Pre-mRNA splicing and other RNA processing events occur co-transcriptionally. High resolution kinetic studies performed in our lab showed splicing-dependent RNA Pol II (RNA polymerase II) pausing near the 3’ splice site of a reporter gene. Pausing requires splicing, as mutations that block splicing lead to loss of pausing, and restoring splicing restores pausing. It was proposed that RNA Pol II pausing may occur at splicing-dependent transcriptional checkpoints. In this study, I aimed to search for splicing helicases that might couple splicing with transcription. The ts alleles prp5-1 and prp16-2 were found to cause transcription defects. These genes encode RNA helicases that were reported to act as fidelity factors during splicing. In vivo RNA labelling and RT-qPCR experiments performed with these temperature-sensitive mutants demonstrated reduced transcription coinciding with the splicing defect at restrictive temperature. Furthermore, RNA Pol II ChIP analysis showed polymerase accumulating over intron-containing genes in both mutants. ChIP analysis using antibodies specific to the phosphorylation status of the CTD (Carboxy Terminal Domain) of RNA Pol II, revealed that the apparently stalled polymerase is hyper-phosphorylated at serine 5. Intriguingly, prp8-R1753K, a ts allele of PRP8, a non-helicase splicing factor mutant also showed reduced nascent RNA synthesis but no RNA Pol II accumulation. To elucidate the reason for the observed RNA Pol II accumulation and to identify a possible splicing-dependent transcriptional checkpoint factor, prp5-1 was investigated further. RNA Pol II ChIP-Seq analysis verified that maximum enrichment genome-wide occurred on introns at restrictive conditions in prp5-1, supporting the earlier observation. Furthermore, the double mutant strain cus2Δprp5-1 abolished the RNA Pol II accumulation observed in prp5-1 at restrictive temperature and restored transcription. Recreating a stalled spliceosome in a U2 mutant strain also showed RNA Pol II accumulation in the presence of Cus2p, as observed in prp5-1. My observations suggest a link between transcription and monitoring of splicing and indicate that Cus2p, a U2 snRNP associated protein, could be a checkpoint factor in transcription prior to pre-spliceosome formation. I speculate that fidelity factors may impose transcriptional checkpoints at different stages of splicing.

Microorganisms are exposed to constantly changing environments, and consequently have evolved mechanisms to rapidly adapt their physiology upon stress imposition. These adaptive responses are coordinated through the rewiring of gene expression via complex networks that control the transcriptional program and the activity of post-transcriptional regulators. Although transcription factors primarily determine which genes are expressed, post-transcriptional regulation has a major role in fine-tuning the dynamics of gene expression. Post-transcriptional control is exerted by RNA-binding proteins and small regulatory RNAs (sRNAs) that bind to mRNA targets and modulate their synthesis, degradation and translation efficiency. In Escherichia coli, sRNAs associated with an RNA chaperone, Hfq, are key post-transcriptional regulators, yet the functions of most of these sRNAs are still unknown. The first step in understanding the roles of sRNAs in regulating gene expression is to identify their targets. To generate transcriptome-wide maps of Hfq-mediated sRNA-mRNA binding, we applied CLASH (cross-linking, ligation and sequencing of hybrids), a method that combines in vivo capture of RNA-RNA interactions, high-throughput sequencing and computational analyses, in E. coli. We uncovered thousands of dynamic growth-stage dependent association of Hfq to sRNAs and mRNAs. The latter confirmed known sRNA-target pairs and identified additional targets for known sRNAs, as well as novel sRNAs in various genomic features along with their targets. These data significantly expand our knowledge of the sRNA-target interaction networks in E.coli. In particular, the Hfq CLASH data indicated 3'-UTRs of mRNAs as major reservoirs of sRNAs, and the utilization of these may be more common than anticipated. Our findings also provide mechanistic insights that ensue from the identification of tens of sRNA-sRNA interactions that point to extensive sponging activity among regulatory RNAs: many sRNAs appear to be able to interact and repress the functions of other base-pairing sRNAs. We validated and highlighted the biological significance of some of the CLASH results by characterizing a 3'-UTR derived sRNA, MdoR (mal-dependent OMP repressor). This sRNA emerges by processing of the last transcript of malEFG polycistron, encoding components of maltose transport system. We found MdoR directly downregulates several major porins, whilst derepressing the maltose-specific porin LamB via destabilization of its inhibitor, MicA, likely by a sponging mechanism. Physiologically, MdoR contributes to the remodelling of envelope composition and links nutrient sensing to envelope stress responses during maltose assimilation. MdoR is a clear example of how cells integrate circuitry through multiple networks as part of their adaptive responses and how the CLASH methodology can help expand our understanding of sRNA-based regulation.

Uridylyl insertion/deletion mRNA editing is essential for mitochondrial gene expression in Trypanosoma brucei and governed by multi-protein complexes called editosomes. The final step in each cycle of this post-transcriptional process is that of re-ligating the edited mRNA fragments. The ~20S RNA editing core complex contains two RNA editing ligases, REL1 and REL2, located, respectively, in a deletion and an insertion subcomplex. While REL1 is clearly essential for RNA editing, REL2 knockdown by RNAi has not resulted in a detectable phenotype. To explain these findings, alternative scenarios have been suggested: (a) REL2 is not functional in vivo; (b) REL1 can function in both insertion and deletion editing, whereas REL2 can only function in insertion editing; (c) REL1 has an additional role in repairing erroneously cleaved mRNAs. To further investigate respective functions of the two RELs this study used three complimentary approaches: (i) genetic complementation with chimeric ligase enzymes, (ii) deep sequencing of RNA editing intermediates after ligase inactivation, and (iii) evolutionary analysis. In vivo expression of two chimeric ligases, providing a REL2 catalytic domain at REL1’s position in the deletion subcomplex and a REL1 catalytic domain at REL2’s position in the insertion subcomplex, did not rescue the growth defect caused by REL1 ablation. Although the results were not fully conclusive they suggest that it is the specific catalytic properties of REL1 rather than its position within the deletion subcomplex that makes it essential. In order to identify in vivo substrates of REL1, specific editing intermediates that accumulated after genetic knockdown of REL1 expression were captured by 5’ linker and deep sequenced using Ion Torrent and Illumina technology. Analyses of such unligated editing intermediates with bespoke bioinformatics tools suggest that REL1 functions in deletion editing as expected, but also in the repair of miscleaved mRNAs, implying a novel role for this ligase. Neither role can be fulfilled by REL2, at least not with sufficient efficiency. Sequencing data also suggest that either REL1 is not involved in ligation of addition editing substrates, or that REL2 in this case can fully compensate for loss of REL1. REL1, REL2 and KREPA3 sequences were subjected to analysis using MEGA5 and the HyPhy package available on the Datamonkey adaptive evolution server. Results indicated that all three editosome genes are under much stronger purifying than diversifying selective forces. In general this selection pressure to conserve protein sequence increased from KREPA3 to REL2 to REL1, suggesting a requirement to maintain catalytic function for both ligases. Taken together, these experiments reveal a novel function for REL1 during RNA editing, providing a rationale for its essentiality. Deductively, the results also suggest REL2, which was previously thought to be non-essential, may still be required by the cell at its position in the addition subcomplex. Evolutionary analysis suggests that the RELs and KREPA3 are under the same evolutionary forces to maintain their respective functions in RNA editing.