Peritrichously flagellated bacteria carry out chemotaxis by modulating the frequency of switching between smooth swimming and tumbling. The tumbling frequency is controlled by a signal transduction cascade in which transmembrane receptors modulate the activity of a histidine kinase CheA that transfers phosphate to its cognate response regulator CheY. The proteins of the chemotaxis signaling cascade are localized to clusters found primarily at the poles of cells. In this work, the localization of the CheZ protein, a phosphatase that dephosphorylates CheY~P, is examined. Using a CheZ-GFP fusion protein, we show that CheZ was localized to the polar receptor patch via interaction with the short form of CheA (CheAS). Aromatic residues of CheZ near one end of the elongated CheZ four-helix bundle were determined to be critical for localization. Aliphatic residues in CheAS were also determined to be critical for CheZ localization to the receptor patch and substitution of these residues conferred a tumble bias to swimming cells. A mechanism of CheZ localization is proposed in which the CheZ apical loop interacts with a binding site formed by dimerization of the P1 domain of CheAS. The potential role of CheZ localization as a means of coordinating the rotation state of peritrichously distributed flagella is discussed.

The organization of biological processes via protein-protein interactions and the subcellular localization of enzymes is believed to be fundamental to many aspects of metabolism. Although this organization has been demonstrated in several systems, the mechanisms by which it is established and regulated are still not well understood. The flavonoid biosynthetic pathway offers a unique system in which to study several important aspects of metabolism. Here we describe a novel toolset of mutant alleles within the flavonoid biosynthetic pathway. In addition, we discuss the use of several of these alleles together with a number of emerging technologies to probe the role of subcellular localization of chalcone synthase, the first committed flavonoid biosynthetic enzyme, on metabolic flux, and to characterize a novel chalcone synthase-interacting protein. The over-expression of this interacting protein induces novel phenotypes that are likely associated with the production or distribution of auxin. Further, interaction analyses between recombinant flavonoid biosynthetic enzymes point to the possibility that post-translational modifications play an important role in promoting interactions. / Ph. D.

Many mitochondrial and chloroplast proteins are encoded in the nucleus and subsequently imported into the organelles via active protein transport systems. While usually highly specific, some proteins are dual-targeted to both organelles. In tobacco (<i>Nicotiana tabacum L.</i>), the cDNA encoding the mitochondrial isoform of NADP+-dependent isocitrate dehydrogenase (NADP+-ICDH) contains two translational ATG start sites, indicating the possibility of two tandem targeting signals. In this work the putative mitochondrial and chloroplastic targeting signals from NADP+-ICDH were fused to a yellow fluorescent protein (YFP) to generate a series of constructs and introduced into tobacco leaves by <i>Agrobacterium</i>-mediated transient transfection. The subsequent sub-cellular locations of the ICDH:YFP fusion proteins were then examined under the confocal microscope. Constructs predicted to be targeted to the chlroplast all localized to the chloroplast. However, this was not the case for constructs that were predicted to be mitochondrial targeted. While some constructs localized to mitochondria, others appeared to be chloroplast localized. This was attributed to an additional 50 amino acid residues of the mature NADP+-ICDH protein which was present in those constructs. In addition, during the process of generating these constructs our sequence analysis indicated a stop codon present at amino acid position 161 of the mature NADP+-ICDH protein from both Xanthi and Petit Havana cultivars of tobacco. This was confirmed by multiple sequencing reactions and created discrepancies with the reported sequence present in the database. The results of this study raise interesting questions with regard to the targeting and processing of NADP+-ICDH.

Many mitochondrial and chloroplast proteins are encoded in the nucleus and subsequently imported into the organelles via active protein transport systems. While usually highly specific, some proteins are dual-targeted to both organelles. In tobacco (<i>Nicotiana tabacum L.</i>), the cDNA encoding the mitochondrial isoform of NADP+-dependent isocitrate dehydrogenase (NADP+-ICDH) contains two translational ATG start sites, indicating the possibility of two tandem targeting signals. In this work the putative mitochondrial and chloroplastic targeting signals from NADP+-ICDH were fused to a yellow fluorescent protein (YFP) to generate a series of constructs and introduced into tobacco leaves by <i>Agrobacterium</i>-mediated transient transfection. The subsequent sub-cellular locations of the ICDH:YFP fusion proteins were then examined under the confocal microscope. Constructs predicted to be targeted to the chlroplast all localized to the chloroplast. However, this was not the case for constructs that were predicted to be mitochondrial targeted. While some constructs localized to mitochondria, others appeared to be chloroplast localized. This was attributed to an additional 50 amino acid residues of the mature NADP+-ICDH protein which was present in those constructs. In addition, during the process of generating these constructs our sequence analysis indicated a stop codon present at amino acid position 161 of the mature NADP+-ICDH protein from both Xanthi and Petit Havana cultivars of tobacco. This was confirmed by multiple sequencing reactions and created discrepancies with the reported sequence present in the database. The results of this study raise interesting questions with regard to the targeting and processing of NADP+-ICDH.

RNAs undergo different posttranscriptional chemical modifications, which affect their structural stability and functional diversity. RNA methylation is a very common type of post-transcriptional modification and is present in all domains of life: Archaea, Eukaryotes and Bacteria. Some of these methylations are catalyzed either by a RNA-protein complex or by stand-alone enzymes. The RNA-protein complex (Ribonucleoprotein complex) is comprised of a small RNA known as the guide RNA (Box C/D RNA) and core proteins (L7Ae, Nop5, and Fibrillarin). Box C/D RNAs contain conserved regions, called box C and box D near their 5’ and 3’ termini, respectively, and their imperfect copies called box C’ and box D’, internally. A short stretch of sequence between these Boxes are known as the guide/spacer regions, as the guide region helps in recruiting and positioning a specific target RNA for modification. Both in Archaea and Eukarya, box C and box D, as well as box C’ and box D’ together can form a structure called a Kink-turn (K-turn) that is characterized by a canonical Watson-Crick base-paired stem on one side, and a non-canonical stem on the other, separated by a 3-nucleotide loop. In Archaea box C’ and D’ can also form a K-loop, where the canonical stem of K-turn is replaced by a loop. Archaeal L7Ae binds first to the K-turn or K-loop and allows the recruitment of other proteins to form the complex. The presence of a unique box C/D RNA of Haloferax volcanii, called sR-tMet has been reported previously to guide the 2’-O-methylation of C34 in elongator pre-tRNAMet. Here we tried to characterize the structure-function relationship of this guide RNA under in vivo conditions. This RNA lacks a conventional K-turn or K-loop at its C’/D’ motif. We have created an H. volcanii strain that has a genomic deletion of sR-tMet. The sR-tMet gene is not essential for H. volcanii but this sR-tMet deleted strain lacks the 2’-O-methylation of C34 of its elongator tRNAMet. Unlike the close sR-tMet homologs (sR8 from Methanocaldococcus jannaschii and sR49 from Pyrococcus abyssi), the Box C’/D’ motif of sR-tMet is neither a K-turn nor a K-loop. The introduction of proper K-loop in the Box C’/D’ motif (sR-tMet with K-loop) abolished its Cm34 modification function in ΔsR-tMet strain. Direct interaction between L7Ae and the K-loop is not an absolute requirement for its function. However, disruption of the G/A and A/G pairing in Box C/D motif and Box D’ suggests the importance of these non-Watson crick base pairings in respect to sR-tMet’s function. Several other mutational studies have revealed that peculiar sR-tMet guide RNA from H. volcanii, behaves more like a Eukaryotic Box C/D RNA (where the K-loop is not required and presence of longer spacer length) than regular Archaeal one. Pseudouridine synthase 10 (Pus10) is the most recently identified Ψ synthase, found only in higher eukaryotes and Archaea. Archaeal Pus10 produces either tRNA Ψ55 or both tRNA Ψ54 and Ψ55 modifications. In Human, its Ψ synthase activity is not yet confirmed and interestingly it has been implicated in apoptosis. Herein for the first time we revealed that this putative RNA Ψ synthase protein, Human Pus10 (HuP10), translocates from the nucleus to the cytoplasm in TRAIL induced apoptosis. This nucleo-cytoplasmic movement of HuP10 occurs through the CRM1 mediated nuclear export pathway and Caspase 3 influences this movement. HuP10 also mediates crosstalk between the extrinsic and intrinsic pathways during TRAIL-induced apoptosis. Other than its involvement in apoptosis, we have also uncovered that HuP10 is involved in regulation of cell proliferation. Depletion (knockdown) of this protein in different cancer cell lines, promotes cell migration and anchorage-independent cell growth in the absence of any apoptotic stimulation.

Stem cells have great potential value for treating a number of diseases and conditions, including diabetes, Parkinson's, and spinal cord injuries. Applying stem cells for therapeutic purposes will require an in-depth understanding of their biology, not only of the genes they express, but also the functions of the proteins encoded by the genes. The goal of the project presented in this thesis was to develop a method for high-throughput analyses of protein localization in mouse stem cells. Localization information can provide insight into the functions and biological roles of proteins. <br /><br /> One means of studying protein localization involves creating proteins with a green fluorescent protein (GFP) reporter gene and analyzing their localization using fluorescence microscopy. The research outlined in this thesis focused on developing a system to create a large number of GFP-tagged proteins by constructing a cDNA?GFP fusion library. This involved exploring methods for optimizing cDNA synthesis, designing a retroviral vector (pBES23) for the expression of cDNA?GFP fusions in mouse stem cells, and constructing a cDNA?GFP fusion library in this vector using R1 mouse embryonic stem cell mRNA. The library constructed was not successfully delivered to target cells for GFP-tagged protein expression; it was therefore not possible to characterize protein localization in mouse stem cells. Suggestions are given as to how the methods used in this thesis might be optimized further.

Doctor of Philosophy / Department of Computing and Information Sciences / Doina Caragea / The large volume of data generated in the recent years has created opportunities for discoveries in various fields. In biology, next generation sequencing technologies determine faster and cheaper the exact order of nucleotides present within a DNA or RNA fragment. This large volume of data requires the use of automated tools to extract information and generate knowledge. Machine learning classification algorithms provide an automated means to annotate data but require some of these data to be manually labeled by human experts, a process that is costly and time consuming. An alternative to labeling data is to use existing labeled data from a related domain, the source domain, if any such data is available, to train a classifier for the domain of interest, the target domain. However, the classification accuracy usually decreases for the domain of interest as the distance between the source and target domains increases. Another alternative is to label some data and complement it with abundant unlabeled data from the same domain, and train a semi-supervised classifier, although the unlabeled data can mislead such classifier. In this work another alternative is considered, domain adaptation, in which the goal is to train an accurate classifier for a domain with limited labeled data and abundant unlabeled data, the target domain, by leveraging labeled data from a related domain, the source domain. Several domain adaptation classifiers are proposed, derived from a supervised discriminative classifier (logistic regression) or a supervised generative classifier (naïve Bayes), and some of the factors that influence their accuracy are studied: features, data used from the source domain, how to incorporate the unlabeled data, and how to combine all available data. The proposed approaches were evaluated on two biological problems -- protein localization and ab initio splice site prediction. The former is motivated by the fact that predicting where a protein is localized provides an indication for its function, whereas the latter is an essential step in gene prediction.

Giardia intestinalis is an anaerobic protozoan pathogen, agent of the disease known as giardiasis. The regulation of gene expression during giardia cell- and life-cycle has been poorly studied so far, with the exception of variable surface proteins, which constitute the immunoprotective coat of the cell. In this diploma thesis, we focus on the possible role of the 3' untranslated region (3'UTR) of mRNA that mediate stability and localization of mRNA transcripts. We use RNA binding proteins of PUF family, which control the function of the target transcripts by their repression, activation or sequestration, to monitor and verify the role of 3'UTRs. These only eukaryotic proteins are highly evolutionarily conserved. Each of them contain highly conserved C-terminal domain, which specificly binds to 3'UTR of mRNAs. We have identified five different PUF proteins in the genome of G. intestinalis (GiPUF), cinfirmed their expression in G. intestinalis trophozoites and located all five proteins in the cytoplasm. GiPUF2, GiPUF3 and GiPUF5 show an additional affinity to the surface of the endoplasmic reticulum. We have identified the C-terminal binding domain in protein sequences of all GiPUF. The most conserved GiPUF4 contain eight binding sites, nearly identical to the binding site of human Pum1 protein,...

Nos organismos eucariotos, aproximadamente 50% das proteínas traduzidas no citoplasma são transportadas para as organelas, onde irão desempenhar suas funções. Com isso, surgiu um intricado sistema de transporte intracelular de proteínas. Nas plantas, a presença de uma segunda organela endossimbionte, o plastídio, tornou este sistema mais complexo e gerou demanda adicional por transporte. Ainda, grande maioria das proteínas mitocondriais e plastidiais são codificadas por genes nucleares e importadas do citosol. O dogma uma proteína-uma localização foi associado ao conceito de um gene-uma proteína na biologia celular. Entretanto, proteínas individuais podem ter mais de uma função, e mais recentemente, proteínas codificadas por um único gene foram identificadas em mais de um compartimento subcelular, o que deu origem ao conceito de duplo direcionamento (DD). Um exemplo bem estudado de DD vem das proteínas da família das aminoacil-tRNA sintetases (aaRS), que participam da síntese protéica ao acoplar o aminoácido ao seu tRNA cognato. Dentre as aaRSs, a glutamil-tRNA sintetase citosólica (GluRS), através de sua extensão N-terminal, parece estar envolvida com outras funções além da tradução. Em Arabidopsis thaliana, há dois genes nucleares que codificam a GluRS, um para uma proteína de duplo direcionamento (DD) e outro para uma proteína citosólica. Resultados recentes em nosso laboratório mostraram que a GluRS citosólica pode estar relacionada ao controle da localização subcelular de proteínas organelares em Arabidopsis. Para verificar um eventual papel desta proteína na localização subcelular de outras proteínas, foram realizados ensaios de duplo-híbrido em levedura, os quais mostraram interação entre a GluRS e a glutamina sintetase (GS) de Arabidopsis thaliana, proteína de DD para mitocôndrias e cloroplastos Esta interação foi confirmada in planta, sendo a sequência da GluRS responsável pela interação localizada na região N-terminal, do resíduo 207 ao 316. Análises filogenéticas apontam que esta região encontra-se ausente nas bactérias e que originou-se provavelmente em Archea, entre 2,6 e 1,8 bilhões de anos. Além disso, observa-se que esta sequência é conservada em fungos, musgos e plantas vaculares, tendo originado-se em Arabidopsis há cerca de 2 bilhões de anos. / In eukaryotic organisms, about 50% of cytoplasmic translated proteins are transported to the organelles, where they can play their roles. Thus, a complex system for intracellular transport was established. In plants, the presence of a second endosymbiont organelle, the plastid, turned this system still more intricated and required an additional transport mechanism. Besides, most of organellar proteins are coded by nuclear genes and imported from the cytosol. The one protein-one localization was associated to the idea of one gene-one protein, which has long been established in molecular biology. However, individual proteins can show more than one function, and recently, proteins coded by one single gene were identified in more than one subcellular compartment, which has originated the concept of dual targeting. One of the most studied example of dual targeted proteins is the aminoacyl-tRNA synthetase (aaRS) family, which are related to protein synthesis by attaching the correct amino acid onto the cognate tRNA molecule. Among the aaRSs, cytosolic glutamyl-tRNA synthetase (GluRS), through its N-terminal extension, seems to be involved in other cellular role beyond translation. In Arabidopsis thaliana, there are two genes encoding GluRS, one for a dual-targeted protein and other for a cytosolic protein. Recent results in our laboratory showed that GluRS interacts with proteins destinated to other organelles, which suggest that this protein might have a role in interfering on protein localization in Arabidopsis. In order to gain some information on the role of this protein in subcellular localization, yeast two-hybrid assays were performed. These studies showed the interaction between GluRS and glutamine synthetase (GS), a mitochondrial and chloroplastic dual-targeted protein. This interaction was confirmed in planta. In addition, the GluRS sequence associated to protein interaction was localized at its N-terminal portion, between the residues 207 316. Phylogenetic analysis revealed that this region is absent in bacteria and it probably arose from Archea between 2.6 and 1.8 billion years ago. Also, this sequence is conserved in fungi, moss and all the green plants investigated. Finally, datation analysis showed that this sequence arose in Arabidopsis between 2 and 1.7 billion years ago.

A regulação da localização de proteínas é um dos aspectos fundamentais na biologia celular vegetal. Os cloroplastos importam mais de 90% de suas proteínas do citosol, portanto, é importante caracterizar os fatores citosólicos que podem estar envolvidos no direcionamento de proteínas para as organelas. Um ensaio de duplohíbrido em leveduras com as proteínas cloroplastidiais HMPPK/TMPPase (TH1) e Glutamina Sintetase (GS) II usados como iscas revelou que a forma citosólica da glutamil-tRNA sintetase - GluRS (At5g26710) de Arabidopsis thaliana interagiu com ambas as proteínas. Estudos de Complementação da Fluorescência Bimolecular (BiFC) confirmaram tais interações in planta. Estudos com deleções na região Nterminal da GluRS mostraram que esta região é responsável pelas interações com HMPPK/TMPPase e GSII. Além disso, seis resíduos de aminoácidos parecem ser cruciais para a interação entre as proteínas. Curiosamente, foi mostrado que a GluRS está envolvida na localização de proteínas em leveduras. A fim de obter mais informações sobre o envolvimento da GluRS ns localização de proteínas nos cloroplastos, foram produzidos plantas de tabaco transgênicas expressando uma proteína quimérica, feita pela fusão do gene codificador da HMPPK/TMPPase, TH1- GFP, e GSII-GFP e posteriormente usados em ensaios de agroinfiltração com RNA de interferência (RNAi) para GluRS. Análises em microscópio confocal mostraram que TH1-GFP e GSII-GFP acumulam no citosol em vez de serem direcionados aos cloroplastos. Neste trabalho, mostramos pela primeira vez que a GluRS está envolvida na localização de proteínas cloroplastidiais em plantas e esse mecanismo é também conservado em Saccharomyces cerevisiae. / Regulation of protein localization is one of the key aspects in plant cell biology. Chloroplasts import more than 90% of their proteins from the cytosol, therefore, it is important to identify and characterize cytosolic factors that might be involved in protein delivery to the organelar envelope. A yeast two-hybrid screen with a chloroplastlocalized HMPPK/TMPPase protein and glutamine synthetase (GS), used as baits, revealed that the cytosolic form of the glutamyl-tRNA synthetase (GluRS) (At5g26710) from Arabidopsis thaliana interacted with both proteins. Bimolecular Fluorescence Complementation (BiFC) studies confirmed such interactions in planta. Deletion studies of GluRS showed that the N-terminal region of the protein is responsible for proteinprotein interactions (PPI) with TH1 and GS. In addition, six amino acid residues appeared to be crucial for PPI. Interestingly, GluRS has been also shown to be involved in regulating protein localization in yeast. In order to gain more information about the involvement of GluRS on protein localization in chloroplasts, we produced transgenic tobacco plants expressing a chimeric protein made by the fusion of TH1- GFP and GSIIGFP and agroinfiltrated with a RNA interference (RNAi) construct against GluRS. Confocal analysis showed that TH1-GFP and GSII-GFP accumulated in the cytosol instead of being targeted to chloroplasts. Here, we show for the same time that GluRS is involved in protein localization in plants and this mechanism is also conserved in Saccharomyces cerevisiae.