Thesis advisor: Welkin E. Johnson / Thesis advisor: Peter G. Clote / Synthetic biology currently is a rapidly emerging discipline, where innovative and interdisciplinary work has led to promising results. Synthetic design of RNA requires novel methods to study and analyze known functional molecules, as well as to generate design candidates that have a high likelihood of being functional. This thesis is primarily focused on the development of novel algorithms for the design of synthetic RNAs. Previous strategies, such as RNAinverse, NUPACK-DESIGN, etc. use heuristic methods, such as adaptive walk, ensemble defect optimization (a form of simulated annealing), genetic algorithms, etc. to generate sequences that minimize specific measures (probability of the target structure, ensemble defect). In contrast, our approach is to generate a large number of sequences whose minimum free energy structure is identical to the target design structure, and subsequently filter with respect to different criteria in order to select the most promising candidates for biochemical validation. In addition, our software must be made accessible and user-friendly, thus allowing researchers from different backgrounds to use our software in their work. Therefore, the work presented in this thesis concerns three areas: Create a potent, versatile and user friendly RNA inverse folding algorithm suitable for the specific requirements of each project, implement tools to analyze the properties that differentiate known functional RNA structures, and use these methods for synthetic design of de-novo functional RNA molecules. / Thesis (PhD) — Boston College, 2016. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Biology.

RNA helicases are found in all kingdoms of life, functioning in all aspects of RNA biology mainly through modulating structures of RNA and ribonucleoprotein (RNP) complex. RNA structures have fundamental impacts on steps in gene expression, including transcription, pre-mRNA processing, and translation. However, the precise roles and regulatory mechanisms of RNA structures in co- and post-transcriptional processes remain elusive. By probing genome-wide RNA structures in vivo, a recent study suggested that ATP-dependent factors, such as RNA helicases, maintain the actively unfolded state of RNAs. Among all RNA helicases, DEAD-box proteins form the largest family in eukaryotes, and have been shown to remodel RNA/RNP structures both in vitro and in vivo. Nevertheless, for the majority of these enzymes, it is largely unclear what RNAs are targeted and where they modulate RNA/RNP structures to regulate co-transcriptional processes. To fill the gap, my research focused on identification of the RNAs and structures targeted by the DEAD-box protein Dbp2 in S. cerevisiae to uncover the cellular processes that Dbp2 is involved in.<br><div><div>My studies revealed a role of Dbp2 in transcriptional termination. Dbp2 binds to ~34% of yeast mRNAs and all snoRNAs, and loss of DBP2 leads to a termination defect as evidenced by RNA polymerase II (RNAPII) accumulation at 3’ ends of these genes. In addition, the binding pattern of Dbp2 in mRNAs is highly similar to Nrd1 and Nab3 in the Nrd1-Nab3-Sen1 (NNS) termination complex, and deletion of DBP2 leads to reduced recruitment of Nrd1 to its target genomic loci. In Dbp2 and NNS targeted 3’ UTRs, RNA structural changes resulted from DBP2 deletion also overlap polyadenylation elements and correlate with inefficient termination, and loss of stable structure in the 3’ UTR bypasses the requirement for Dbp2. These findings lead to a model that Dbp2 promotes efficient termination of transcription through RNA structure remodeling.</div><div>Interestingly, my research also revealed the requirement of DBP2 for efficient splicing, as loss of DBP2 leads to accumulation of unspliced pre-mRNAs. Moreover, this function is dependent on the helicase activity of Dbp2. Further studies are needed to characterize the molecular mechanism of how Dbp2 facilitates splicing in cells. Overall, my research demonstrated that DEAD-box RNA helicases remodel mRNA structure in vivo and that structural alteration can be essential for proper gene expression.</div></div>

Les ARNs long non codants (ARNlnc) jouent un rôle clé dans les processus cellulaires vitaux, notamment le remodelage de la chromatine, la réparation de l'ADN et la traduction. Cependant, la taille et la complexité des ARNlnc présentent des défis sans précédent pour les études moléculaires mécanistiques, de sorte qu'il s'est avéré difficile jusqu'à présent de relier l'information structurelle à la fonction biologique pour les ARNlnc.Le gène 3 humain exprimé maternellement (de l’anglais "maternally expressed gene 3", MEG3), est un ARNlnc abondant, soumis à empreinte parentale et épissé alternativement. Pendant l'embryogenèse, MEG3 contrôle les protéines Polycomb, régulant la différenciation cellulaire, et dans les cellules adultes, MEG3 contrôle p53, régulant la réponse cellulaire aux stress environnementaux. Dans les cellules cancéreuses, MEG3 est régulé négativement, mais la surexpression ectopique de MEG3 réduit la prolifération incontrôlée, ce qui prouve que MEG3 agit comme un suppresseur de tumeur. Les données suggèrent que les fonctions de MEG3 pourraient être régulées par la structure de MEG3. Par exemple, on pense que MEG3 se lie directement aux protéines p53 et Polycomb. De plus, les différents variants d'épissage de MEG3, qui comprennent différents exons et possèdent ainsi des structures potentiellement différentes, présentent des fonctions différentes. Enfin, la mutagenèse par délétion, basée sur une structure de MEG3 prédit in silico, a permis d’identifier un motif MEG3 supposé structuré impliqué dans l'activation de p53. Cependant, au début de mes travaux, la structure expérimentale de MEG3 était inconnue.Pour comprendre la structure et la fonction de MEG3, j'ai utilisé des sondes chimiques in vitro et in vivo pour déterminer la structure secondaire de deux variants humains de MEG3 qui diffèrent par leurs niveaux d'activation de p53. À l'aide d'essais fonctionnels dans les cellules et de mutagenèse, j'ai systématiquement analysé la structure de MEG3 et identifié le noyau activant p53 dans deux domaines (D2 et D3) qui sont conservés structuralement dans les variants humains et conservés dans l’évolution chez les mammifères. Dans D2-D3, les régions structurales les plus importantes sont les hélices H11 et H27, car dans ces régions, j’ai pu supprimer l'activation de p53 grâce à des mutations ponctuelles, un degré de précision jamais atteint pour les autres ARNlnc jusqu’ici. J'ai découvert de manière surprenante que H11 et H27 sont reliés par des boucles connectées l’une à l’autre (de l’anglais "kissing loops") et j'ai confirmé l'importance fonctionnelle de ces interactions de structure tertiaire à longue distance par mutagenèse compensatoire. Allant au-delà de l’état de l’art, j'ai donc essayé de visualiser la structure 3D d’une isoforme de MEG3 longue de 1595 nucléotides, par diffusion de rayons X à petit angle (SAXS), microscopie électronique (EM) et microscopie à force atomique (AFM). Alors que le SAXS et l’EM sont limités par des défis techniques actuellement insurmontables, l’imagerie par AFM m’a permis d’obtenir la première structure 3D à basse résolution de MEG3 et de révéler son échafaudage tertiaire compact et globulaire. Plus remarquable encore, les mêmes mutations qui perturbent la connexion entre les «boucles» H11-H27 et qui inhibent la fonction de MEG3, perturbent aussi la structure 3D de cet ARNlnc, fournissant ainsi le premier lien direct entre la structure 3D et la fonction biologique pour un ARNlnc.Sur la base de mes découvertes, je peux donc proposer un mécanisme de l’activation de p53 basé sur la structure de MEG3, avec des implications importantes pour la compréhension de la cancérogenèse. Plus généralement, mes travaux prouvent que les relations structure-fonction des ARNlnc peuvent être disséquées avec une grande précision et ouvrent la voie à des études analogues visant à obtenir des informations mécanistes pour de nombreux autres ARNlnc d’importance médicale. / Long non-coding RNAs (lncRNAs) are key players in vital cellular processes, including chromatin remodelling, DNA repair and translation. However, the size and complexity of lncRNAs present unprecedented challenges for mechanistic molecular studies, so that connecting structural information with biological function for lncRNAs has proven difficult so far.Human maternally expressed gene 3 (MEG3) is an abundant, imprinted, alternatively-spliced lncRNA. During embryogenesis MEG3 controls Polycomb proteins, regulating cell differentiation, and in adult cells MEG3 controls p53, regulating the cellular response to environmental stresses. In cancerous cells, MEG3 is downregulated, but ectopic overexpression of MEG3 reduces uncontrolled proliferation, proving that MEG3 acts as a tumour suppressor. Evidence suggests that MEG3 functions may be regulated by the MEG3 structure. For instance, MEG3 is thought to bind p53 and Polycomb proteins directly. Moreover, different MEG3 splice variants, which comprise different exons and thus possess potentially different structures, display different functions. Finally, deletion mutagenesis based on a MEG3 structure predicted in silico identified a putatively-structured MEG3 motif involved in p53 activation. However, at the beginning of my work, the experimental structure of MEG3 was unknown.To understand the MEG3 structure and function, I used chemical probing in vitro and in vivo to determine the secondary structure maps of two human MEG3 variants that differ in their p53 activation levels. Using functional assays in cells and mutagenesis, I systematically scanned the MEG3 structure and identified the p53-activating core in two domains (D2 and D3) that are structurally conserved across human variants and evolutionarily conserved across mammals. In D2-D3, the most important structural regions are helices H11 and H27, because in these regions I could tune p53 activation even by point mutations, a degree of precision never achieved for any other lncRNA to date. I surprisingly discovered that H11 and H27 are connected by “kissing loops”, and I confirmed the functional importance of these long-range tertiary structure interactions by compensatory mutagenesis. Going beyond state-of-the-art, I thus attempted to visualize the 3D structure of a 1595-nucleotide long MEG3 isoform by small angle X-ray scattering (SAXS), electron microscopy (EM), and atomic force microscopy (AFM). While SAXS and EM are limited by currently-insurmountable technical challenges, single particle imaging by AFM allowed me to obtain the first low resolution 3D structure of MEG3 and reveal its compact, globular tertiary scaffold. Most remarkably, functionally-disrupting mutations that break the H11-H27 “kissing loops” disrupt such MEG3 scaffold, providing the first direct connection between 3D structure and biological function for an lncRNA.Based on my discoveries, I can therefore propose a structure-based mechanism for p53 activation by human MEG3, with important implications in understanding carcinogenesis. More broadly, my work serves as proof-of-concept that lncRNA structure-function relationships can be dissected with high precision and opens the field to analogous studies aimed to gain mechanistic insights into many other medically-relevant lncRNAs.

Chemical modification and limited enzymatic hydrolysis are powerful methods to obtain detailed information on the structure and dynamics of RNAs in solution. In the work presented here I have taken advantage of the properties of the divalent metal ion lead(II) to establish it as a new probe for investigating the structure of RNA in vivo. Besides highly specific lead(II)-induced cleavage due to the presence of tight metal ion binding sites, lead(II) is known to cleave RNA within single-stranded regions, loops and bulges. The detailed structural data obtained with three different RNAs: tmRNA, CopT, and the leader region of the ompF mRNA, show that lead(II) has great potential for in vivo studies of RNA structure. In P. fluorescens, the activity and stability of RsmY, a small regulatory RNA, was shown to be strongly dependent on repeated GGA motifs in single-stranded regions. In vivo lead(II) probing essentially confirmed predicted secondary structures and also indicated binding to a protein, RsmA. The potential in using lead(II) for mapping protein binding sites on RNAs was shown for the interaction between E. coli tmRNA and the SmpB protein. In vivo and in vitro data show protections in the tRNA-like domain of tmRNA due to binding to the SmpB protein, indicating that the SmpB protein is associated with the majority of tmRNA in the cell. Furthermore, the overall conformation/ structure of E. coli RNase P was analyzed by probing the native structure of M1 RNA in vivo with lead(II). The observed cleavages suggests that M1 RNA is present in two main conformations in the cell, one being characteristic of free RNase P, and one of an RNase P-tRNA complex. The results also indicate that the C5 protein subunit has only minor effects on the overall structure of the RNA subunit.

RNA backbone is biologically important with many roles in reactions and interactions, but has historically been a challenge in structural determination. It has many atoms and torsions to place, and often there is less data on it than one might wish. This problem leads to both random and systematic error, producing noise in an already high-dimensional and complex distribution to further complicate data-driven analysis. With the advent of the ribosomal subunit structures published in 2000, large RNA structures at good resolution, it became possible to apply the Richardson laboratory's quality-filtering, visualization, and analysis techniques to RNA and develop new tools for RNA as well. A first set of 42 RNA backbone rotamers was identified, developed, and published in 2003; it has since been thoroughly overhauled in conjunction with the backbone group of the RNA Ontology Consortium to combine the strengths of different approaches, incorporate new data, and produce a consensus set of 46 conformers. Meanwhile, extensive work has taken place on developing validation and remodeling tools to correct and improve existing structures as well as to assist in initial fitting. The use of base-phosphate perpendicular distances to identify sugar pucker has proven very useful in both hand-refitting and the semi-automated process of using RNABC (RNA Backbone Correction), a program developed in conjunction with Dr. Jack Snoeyink's laboratory. The guanine riboswitch structure ur0039/1U8D, by Dr. Rob Batey's laboratory, has been collaboratively refit and rerefined as a successful test case of the utility of these tools and techniques. Their testing and development will continue, and they are expected to help to improve RNA structure determination in both ease and quality. / Dissertation

The biological function of RNA is often linked to an ability to adopt one or more mutually exclusive conformational states or isomers, a characteristic that distinguishes this biomolecule from proteins. Two examples of conformationally inconverting RNAs were structurally investigated. The first is found in the 3' untranslated region (UTR) of the coronavirus mouse hepatitis virus (MHV). A proposed molecular switch between mutually exclusive stable stem loop and pseudoknot conformations was investigated using thermal unfolding methods, NMR spectroscopy, sedimentation velocity ultracentrifugation and fluorescence resonance energy transfer (FRET) spectroscopy. Utilizing a "divide and conquer" approach we establish that the independent subdomains are folded as predicted by the proposed model and that a pseudoknotted conformation is accessible. Using the subdomains as spectral markers for the investigation of the intact 3' UTR RNA, we show that the 3' UTR is indeed a superposition of a double stem conformation and a pseudoknotted conformation in the presence of KCl and MgCl2. In the absence of added salt however, the 3' UTR adopts exclusively the double stem conformation. Analysis of the pseudoknotted stem reveals only a marginally stable folded state (deltaG25 = 0.5 kcal mol-1, tm = 31 oC) which makes it likely that a viral or host encoded protein(s) is required to stabilize the pseudoknotted conformation. A second conformationally interconverting RNA system investigated is an RNA element that stimulates -1 programmed ribosomal frameshifting in the human Ma3 gene. Structural analysis of the frameshifting element reveals a dynamic equilibrium between a functionally inactive double stem loop conformation and the active pseudoknotted conformation. Thermal melting and NMR spectroscopy reveal that the double stem loop is the predominant conformation in the absence of added KCl or MgCl2. The addition of KCl and MgCl2 results in the formation of a pseudoknot conformation. This conformation is dominant in solution only when the competing double stem loop conformation is abrogated by mutation. Functional studies of the Ma3 pseudoknot reveal that abrogation of double stem conformation increases frameshift stimulation by 2-fold and indicates that the pseudoknot is the active conformation.

This work investigates pathways of conformational transitions in ubiquitous RNA structural motifs. In our lab, we have developed multi-scale structural datamining techniques for identification of three-dimensional structural patterns in high-resolution crystal structures of globular RNA. I have applied these techniques to identify variations in the conformations of RNA double-helices and tetraloops. The datamined structural information is used to propose reaction coordinates for conformational transitions involved in double-strand helix propagation and tetraloop folding in RNA. I have also presented an algorithm to identify stacked RNA bases. In this work, experimentally derived thermodynamic evaluation of the conformations has been used to as an additional parameter to add detail to RNA structural transitions. RNA conformational transitions help control processes in small systems such as riboswitches and in large systems such as ribosomes. Adopting functional conformations by globular RNA during a folding process also involves structural transitions. RNA double-helices and tetraloops are common, ubiquitous structural motifs in globular RNA that independently fold in to a thermodynamically stable conformation. Folding models for these motifs are proposed in this work with probable intermediates ordered along the reaction coordinates. We hypothesize that frequently observed structural states in crystals structures are analogous in conformation to stable thermodynamic â on-pathwayâ folded states. Conversely, we hypothesize that conformations that are rarely observed are improbable folding intermediates, i.e., these conformational states are â off-pathwayâ states. In general on-pathway states are assumed to be thermodynamically more stable than off-pathway states, with the exception of kinetic traps. Structural datamining shows that double helices in RNA may propagate by the â stack-ratchetâ mechanism proposed here instead of the commonly accepted zipper mechanism. Mechanistic models for RNA tetraloop folding have been proposed and validated with experimentally derived thermodynamic data. The extent of stacking between bases in RNA is variable, indicating that stacking may not be a two-state phenomenon. A novel algorithm to define and identify stacked bases at atomic resolution has also been presented in this work.

We describe the use of a tensor mode-1 higher-order singular value decomposition (HOSVD) in the analyses of alignments of 16S and 23S ribosomal RNA (rRNA) sequences, each encoded in a cuboid of frequencies of nucleotides across positions and organisms. This mode-1 HOSVD separates the data cuboids into combinations of patterns of nucleotide frequency variation across the positions and organisms, i.e., "eigenorganisms"' and corresponding nucleotide-specific segments of "eigenpositions," respectively, independent of a-priori knowledge of the taxonomic groups and their relationships, or the rRNA structures. We show that this mode-1 HOSVD provides a mathematical framework for modeling the sequence alignments where the mathematical variables, i.e., the significant eigenpositions and eigenorganisms, are consistent with current biological understanding of the 16S and 23S rRNAs. First, the significant eigenpositions identify multiple relations of similarity and dissimilarity among the taxonomic groups, some known and some previously unknown. Second, the corresponding eigenorganisms identify positions of nucleotides exclusively conserved within the corresponding taxonomic groups, but not among them, that map out entire substructures inserted or deleted within one taxonomic group relative to another. These positions are also enriched in adenosines that are unpaired in the rRNA secondary structure, the majority of which participate in tertiary structure interactions, and some also map to the same substructures. This demonstrates that an organism's evolutionary pathway is correlated and possibly also causally coordinated with insertions or deletions of entire rRNA substructures and unpaired adenosines, i.e., structural motifs which are involved in rRNA folding and function. Third, this mode-1 HOSVD reveals two previously unknown subgenic relationships of convergence and divergence between the Archaea and Microsporidia, that might correspond to two evolutionary pathways, in both the 16S and 23S rRNA alignments. This demonstrates that even on the level of a single rRNA molecule, an organism's evolutionary pathway is composed of different types of changes in structure in reaction to multiple concurrent evolutionary forces. / text

Many functional RNAs are required to fold into specific three-dimensional structures. A fundamental property of RNA is that its secondary structure and even some tertiary contacts are highly stable, which gives rise to independent modular RNA motifs and makes RNAs prone to adopting misfolded intermediates. Consequently, in addition to stabilizing the native structure relative to the unfolded species (defined here as stability), RNAs are faced with the challenge of stabilizing the native structure relative to alternative structures (defined as structural specificity). How RNAs have evolved to overcome these challenges is incompletely understood. Self-splicing group I introns have been used to study RNA structure and folding for decades. Among them, the Tetrahymena intron was the first discovered and has been studied extensively. In this work, we found that a version of the intron that was generated by in vitro selection for enhanced stability also displayed enhanced specificity against a stable misfolded structure that is globally similar to the native state, despite the absence of selective pressure to increase the energy gap between these structures. Further dissection suggests that the increased specificity against misfolding arises from two point mutations, which strengthen a local tertiary contact network that apparently cannot form in the misfolded conformation. Our results suggest that the structural rigidity and intricate networks of contacts inherent to structured RNAs can allow them to evolve exquisite structural specificity without explicit negative selection, even against closely-related alternative structures. To explore further how RNAs gain stability from intricate architectures, we examined a novel group I intron from red algae (Bangia). Biochemical methods and computational modeling suggest that this intron possesses general motifs of group IC1 introns but also forms an atypical tertiary contact, which has been reported previously in other subgroups and helps position the reactive helix at the active site. In the Bangia intron, the partners have been swapped relative to known group I RNAs that include this contact. This result underscores the modular nature of RNA motifs and provides insight into how structured RNAs can arrange helices and contacts in multiple ways to achieve and stabilize functional structures. / text

abstract: Telomerase is a specialized enzyme that adds telomeric DNA repeats to the chromosome ends to counterbalance the progressive telomere shortening over cell divisions. It has two essential core components, a catalytic telomerase reverse transcriptase protein (TERT), and a telomerase RNA (TR). TERT synthesizes telomeric DNA by reverse transcribing a short template sequence in TR. Unlike TERT, TR is extremely divergent in size, sequence and structure and has only been identified in three evolutionarily distant groups. The lack of knowledge on TR from important model organisms has been a roadblock for vigorous studies on telomerase regulation. To address this issue, a novel in vitro system combining deep-sequencing and bioinformatics search was developed to discover TR from new phylogenetic groups. The system has been validated by the successful identification of TR from echinoderm purple sea urchin Strongylocentrotus purpuratus. The sea urchin TR (spTR) is the first invertebrate TR that has been identified and can serve as a model for understanding how the vertebrate TR evolved with vertebrate-specific traits. By using phylogenetic comparative analysis, the secondary structure of spTR was determined. The spTR secondary structure reveals unique sea urchin specific structure elements as well as homologous structural features shared by TR from other organisms. This study enhanced the understanding of telomerase mechanism and the evolution of telomerase RNP. The system that was used to identity telomerase RNA can be employed for the discovery of other TR as well as the discovery of novel RNA from other RNP complex. / Dissertation/Thesis / Ph.D. Biochemistry 2011