Several lipopeptidyl nucleoside antibiotics that inhibit bacterial translocase I (MraY) involved in peptidoglycan cell wall biosynthesis contain an aminoribosyl moiety, an unusual sugar appendage in natural products. A-90289 and muraminomicin are the two representative antibiotics that belong to this family. Bioinformatic analysis of the biosynthetic A-90289 gene clusters revealed that five enzymes are likely involved in the assembly and attachment of the aminoribosyl unit. These enzymes of A-90289 are functionally assigned by in vitro characterization. The results reveal a unique ribosylation pathway that highlighted by uridine-5′-monophosphate as the source of the sugar, a phosphorylase strategy to generate a sugar-1-phosphate, and a primary amine-requiring nucleotidylyltransferase that generates the NDP-sugar donor. Muraminomicin, which has a structure similar to A-90289, holds the distinction in that both ribose units are 2-deoxy sugars. The biosynthetic gene cluster of muraminomicin has been identified, cloned and sequenced, and bioinformatic analysis revealed a minimum of 24 open reading frames putatively involved in the biosynthesis, resistance, and regulation of muraminomicin. Similar to the A-90289 pathway, fives enzymes are still likely involved in the assembly of the 2,5-dideoxy-5-aminoribose saccharide unit, and two are now functionally assigned and characterized: Mra20, a 5′-amino-2′,5′-dideoxyuridine phosphorylase and Mra23, a UTP:5-amino-2,5-dideoxy-α-D-ribose-1-phosphate uridylyltransferase. The cumulative results are consistent with the incorporation of the ribosyl appendage of muraminomicin via the archetypical sugar biosynthetic pathway that parallels A-90289 biosynthesis

La réinitiation de la traduction est un mécanisme permettant de traduire des ORF qui sont présents dans la région leader de différents ARNm cellulaires (uORF). La majorité des cas de réinitiation de la traduction chez les eucaryotes concerne des uORF de petite taille. Des stratégies alternatives ont été développées, entre autres par les virus, afin de réinitier la traduction après un long uORF. Le virus de la mosaïque du chou-fleur (CaMV) exprime un ARNm polycistronique codant la totalité des protéines virales. L’une d’entre elle, la protéine TAV (TransActivateur/Viroplasmine) est un facteur essentiel qui rend possible la réinitiation de la traduction après de longs ORF et qui, de plus, active la protéine kinase TOR. La sous-unité h du facteur d’initiation de la traduction eIF3, requise pour promouvoir la reinitiation après un petit ORF chez les plantes, a été identifiée comme étant une nouvelle cible de phosphorylation de la voie de signalisation de TOR. L’objectif principal de ma thèse a été d’élucider la fonction de la protéine eIF3h dans la réinitiation après un petit ORF ainsi que dans la réinitiation de la traduction, assurée par TAV, après un long ORF. Nous avons exploité les lignées transgéniques eif3h-1 d’Arabidopsis exprimant la protéine eif3h tronquée de son extrémité C-terminale, qui sont déficientes pour la réinitiation mais pas pour l’initiation de la traduction. Nous avons montré que la phosphorylation de eIF3h est essentielle pour stabiliser eIF3 au niveau des ribosomes durant l’élongation, ce qui favorise la ré-acquisition par le ribosome de facteurs nécessaires à la réinitiation de la traduction, et que la délétion de sa région Ct abolit son intégration dans le complexe eIF3. De plus, nous avons montré que eIF3h, la cible de la voie de signalisation de TOR, interagit avec S6K1. Des protoplastes préparés à partir des plantes mutantes eif3h-1 sont incapables de promouvoir la réinitiation après de longs ORF en présence de TAV. La surexpression de eIF3h, indifféremment de son état de phosphorylation, est indispensable pour restaurer la reinitiation assurée par TAV dans les protoplastes eif3h-1. Par ailleurs, les plantes eif3h-1 déficientes dans la réinitiation, sont résistantes à l’infection par le CaMV démontrant l’importance de eIF3h pour la réplication du CaMV. En revanche, ces plantes eif3h-1 peuvent être infectées par d’autres virus dont la traduction de l’ARN génomique est coiffe- ou IRES-dépendante. Ainsi, nos résultats suggèrent que eIF3h est un facteur de reinitiation important aussi bien pour la reinitiation après un petit qu’après un long ORF (controlée par TAV), et que TAV exploite cette machinerie cellulaire, et plus particulièrement TOR et eIF3h, pour exprimer ses propres protéines par réinitiation de la traduction. / Translation of mRNAs that harbor upstream open reading frames (uORFs) within their leader regions operates via a reinitiation mechanism. In plants, reinitiation is up regulated by the target of rapamycin (TOR) signaling via phosphorylation of the subunit h of initiation factor 3 (eIF3). The eif3h-1 mutant expressing the C-terminally truncated eIF3h while maintaining high translation initiation efficiency is not active in reinitiation. Cauliflower mosaic virus (CaMV) pregenomic polycistronic RNA is translated via an exceptional mechanism of reinitiation after long ORF translation under control of CaMV protein TAV, which ensures activation of TOR. To find the link between underlying mechanisms, we examined eIF3h function in cellular and viral context. Here we show that eIF3h, if phosphorylated, has a role in recruitment of eIF3 into actively translating ribosomes that is a prerequisite for formation of reinitiation-competent ribosomal complexes. C-terminal truncation of eIF3h abolished its integration into the eIF3 complex and eIF3 loading on polysomes as manifested by the eIF3 core subunit c. We also show that eIF3h as a putative target of TOR/S6K1 binds S6K1 in vitro. eIF3h phosphorylation is not required for eIF3 complex formation. We demonstrated that eIF3h is essential for TAV to activate reinitiation after long ORF translation. Protoplasts derived from eif3h-1 mutant failed to support TAV function in reinitiation, which is restored only upon overexpression of recombinant eIF3h indifferent to its phosphorylation status. eif3h-1 mutant defective in reinitiation was found resistant to CaMV infection suggesting that eIF3h is critical for virus amplification. In contrast, viruses that evolve translation initiation dependent on either cap or the internal ribosome entry site infect reinitiation deficient mutant. Thus, we conclude that TAV exploits the basic cell reinitiation machinery, particularly TOR and eIF3h, to overcome cellular barriers to reinitiation after long ORF translation.

Genetically modified organisms (GMOs) are confined physically in order to contain their spread in nature and to minimise chances of horizontal gene transfer. However, with the potential that GMOs hold as cheap, reliable and efficient micro-machines, their eventual uncontrolled release into the wider space is becoming more likely. Indeed, their application as environmental sensors is largely increasing. Nevertheless, the field of synthetic biology may also afford solutions to the problem. A major potential application of GMOs is the delivery of antigens to human and animal hosts, through the utilization of live, engineered microbes. Recombinant technology is promising for several reasons including their capacity to be less reactogenic, more potent, safer and genetically definable. Also, they have the potential to provide protection against multiple targets simultaneously, are relatively inexpensive and can be eradicated with antibiotics, as the need arises. Besides, delivery of vaccines to mucosal surfaces is more efficient. Mutant Salmonella expressing heterologous antigens have been shown to induce protection against a variety of pathogens. Nevertheless, limited containment systems are available that can be applicable for bacterial antigen carriers. This project aims to design safeguards for the bacterial antigen delivery systems that limit ORF translatability and self-inactivates/destructs upon exit from the host. In this work, double quadruplet codons were suppressed by orthogonal tRNAs, providing a barrier for gene translation in the recipient cells when antigen is horizontally transferred. Furthermore, three kill switches were designed that are activated by a decrease in temperature from 37 °C. First, Sau3AI endonuclease was activated by protein self-splicing at low temperature mediated by Mtu recA intein. The activation of the endonuclease led to three-fold logarithmic decrease in the number of viable cells within two hours of gene expression. Second, RNA-dependent activation of RNase 7 showed a reduction in the number of viable cells at low temperature of three logarithmic folds. RNase 7 was controlled by the cspA 5’UTR, which sequesters ribosome binding site at 37 °C and allows translation at low temperature. Third, CspA 5’UTR was shown to regulate expression of TEV protease at 37 °C and low temperature. This led to bacterial cellular inhibition within two hours of TEV induction and five-fold logarithmic reduction in the number of viable cells at low temperature. In addition, for the first time and contrary to previous studies, the TEV protease was shown to inhibit cellular growth. It was also shown that biofilm formation was drastically impaired by the TEV activity. The three killing switches and the quadruplet translation system are poised to function as robust safeguards for bacterial antigen delivery systems.

<p>Data science uses the latest techniques in statistics and machine learning to extract insights from data. With the increasing amount of protein data, a number of novel research approaches have become feasible.</p><p>Micropeptides are an emerging field in the protein universe. They are small proteins with <= 100 amino acid residues (aa) and are translated from small open reading frames (sORFs) of <= 303 base pairs (bp). Traditionally, their existence was ignored because of the technical difficulties in isolating them. With technological advances, a growing number of micropeptides have been characterized and shown to play vital roles in many biological processes. Yet, we lack bioinformatics methods for predicting them directly from DNA sequences, which could substantially facilitate research in this field with minimal cost. With the increasing amount of data, developing new methods to address this need becomes possible. We therefore developed MiPepid, a machine-learning-based method specifically designed for predicting micropeptides from DNA sequences by curating a high-quality dataset and by training MiPepid using logistic regression with 4-mer features. MiPepid performed exceptionally well on holdout test sets and performed much better than existing methods. MiPepid is available for downloading, easy to use, and runs sufficiently fast.</p><p>Long noncoding RNAs (LncRNAs) are transcripts of > 200 bp and does not encode a protein. Contrary to their “noncoding” definition, an increasing number of lncRNAs have been found to be translated into functional micropeptides. Therefore, whether most lncRNAs are translated is an open question of great significance. To address this question, by harnessing the availability of large-scale human variation data, we have explored the relationships between lncRNAs, micropeptides, and canonical regular proteins (> 100 aa) from the perspective of genetic variation, which has long been used to study natural selection to infer functional relevance. Through rigorous statistical analyses, we find that lncRNAs share a similar genetic variation profile with proteins regarding single nucleotide polymorphism (SNP) density, SNP spectrum, enrichment of rare SNPs, etc., suggesting lncRNAs are under similar negative selection strength with proteins. Our study revealed similarities between micropeptides, lncRNAs, and canonical proteins and is the first attempt to explore the relationships between the three groups from a genetic variation perspective.</p><p>Deep learning has been tremendously successful in 2D image recognition. Protein binding ligand prediction is fundamental topic in protein research as most proteins bind ligands to function. Proteins are 3D structures and can be considered as 3D images. Prediction of binding ligands of proteins can then be converted to a 3D image classification problem. In addition, a large number of protein structure data are available now. We therefore utilized deep learning to predict protein binding ligands by designing a 3D convolutional neural network from scratch and by building a large 3D image dataset of protein structures. The trained model achieved an average F1 score of over 0.8 across 151 classes on the holdout test set. Compared to existing methods, our model performed better. In summary, we showed the feasibility of deploying deep learning in protein structure research.</p><p>In conclusion, by exploring various edges of the protein universe from the perspective of data science, we showed that the increasing amount of data and the advancement of data science methods made it possible to address a wide variety of pressing biological questions. We showed that for a successful data science study, the three components – goal, data, method – all of them are indispensable. We provided three successful data science studies: the careful data cleaning and selection of machine learning algorithm lead to the development of MiPepid that fits the urgent need of a micropeptide prediction method; identifying the question and exploring it from a different angle lead to the key insight that lncRNAs resemble micropeptides; applying deep learning to protein structure data lead to a new approach to the long-standing question of protein-ligand binding. The three studies serve as excellent examples in solving a wide range of data science problems with a variety of issues.</p>

Les mitochondries, organelles d’origine bactérienne, sont trouvées dans les cellules de presque tous les organismes eucaryotes. Elles exercent des rôles centraux dans les fonctions cellulaires tels que la production d’énergie, la signalisation cellulaire et l’apoptose et ont aussi un impact sur le vieillissement ainsi que certains cancers et maladies neurodégénératives. Chez l’humain et les mammifères en général, le génome mitochondrial est une molécule d’ADN double brin circulaire composée de 37 gènes. Seulement 13 de ces gènes codent des protéines mitochondriales et les 24 autres produisent 22 ARNt (ARN de transfert) et 2 ARNr (ARN ribosomal) qui sont nécessaires à la traduction des 13 protéines mitochondriales. L’ADNmt (ADN mitochondrial) étant très compact, ceci suggère qu’il y a peu de possibilités pour des nouveautés évolutives. Cependant, de récentes recherches ont permis de révéler la présence de près d’une dizaine de petits ORF (cadres de lecture ouverts) fonctionnels à l’intérieur des gènes mitochondriaux 12S ARNr et 16S ARNr. Ceci remet en question la complexité du génome mitochondrial et montre que son potentiel codant a été sous-estimé. Une analyse approfondie du génome mitochondrial humain a révélé la présence de 227 séquences potentiellement traduites en protéines mitochondriales à travers l’ensemble du génome. Dans cette étude, nous avons sélectionné 9 de ces 227 séquences afin de déterminer si effectivement, elles produisent un peptide identifiable. Pour ce faire, des expériences d’immunobuvardage, d’immunofluorescence et d’immunoprécipitation ont été réalisées sur des cellules HeLa et des cellules HEK293T. Ces expériences ont permis d’identifier une protéine mitochondriale alternative nommée MTALTND4 dont la séquence codante est trouvée à l’intérieur du gène nd4, dans un cadre de lecture alternatif. MTALTND4 est traduite dans la mitochondrie et peut être exportée dans le cytoplasme ainsi qu’à l’extérieur de la cellule puisqu’elle a été retrouvée dans le plasma humain. Bien que la fonction de cette protéine n’ait pas encore été confirmée, des résultats préliminaires indiquent qu’elle a un impact sur la respiration cellulaire. MTALTND4 diminue la respiration mitochondriale et nos résultats suggèrent que son action serait induite par l’hypoxie. La découverte de ce nouveau gène mitochondrial humain confirme que le potentiel codant du génome mitochondrial est beaucoup plus vaste que ce que nous croyions. Il existe fort probablement encore plusieurs autres protéines mitochondriales dont les effets pourraient se révéler d’une grande importance. En effet, plusieurs des protéines dérivées du génome mitochondrial découvertes à ce jour ont des impacts majeurs au niveau du métabolisme et pourraient agir en tant que molécules thérapeutiques importantes. Nos résultats amènent à repenser l’évolution et les pressions de sélection exercées sur le génome mitochondrial et ouvrent la porte à de nombreuses recherches futures qui permettront de re-caractériser le génome mitochondrial et d’avoir une compréhension encore plus approfondie du rôle des mitochondries dans les fonctions cellulaires. / Mitochondria, organelles of bacterial origin, are found in almost every eukaryotic organism and play a central role in cellular functions such as energy production, cellular signaling and apoptosis and are also known to have an impact on aging, certain cancers and neurodegenerative diseases. In humans and mammals in general, the mitochondrial DNA is a small double-stranded circular molecule coding for only 37 genes. Only 13 of them code for mitochondrial proteins and the other 24 genes produce 22 tRNAs (transfer RNA) and 2 rRNAs (ribosomal RNA) necessary for the translation of the 13 protein coding genes. The extremely compact nature of mtDNA (mitochondrial DNA) suggests that there is little room for evolutionary novelties. However, recent research revealed the presence of about ten small functional ORFs inside the mitochondrial genes 12S rRNA and 16S rRNA. This calls into question the complexity of the mitochondrial genome and shows that its coding potential has been greatly underestimated. A thorough examination of the human mitochondrial genome revealed the presence of 227 sequences potentially translated into mitochondrial alternative proteins across the entire genome. In this study, we selected 9 of the 227 sequences to determine if they indeed produce identifiable peptides. This was done by immunoblotting, immunofluorescence and immunoprecipitation experiments on HeLa and HEK293T cells. These experiments allowed us to identify one alternative protein named MTALTND4 whose coding sequence is found inside the nd4 gene, in an alternative sequence. MTALTND4 is translated inside the mitochondria and can be exported in the cytoplasm as well as outside the cell since it has been found in human plasma. Although the function of this protein has not yet been confirmed preliminary results indicate its impact on cellular respiration. MTALTND4 decreases mitochondrial respiration and our results suggest that its action could be induced by hypoxia. The discovery of this new human mitochondrial gene confirms that the coding potential of the mitochondrial genome is much larger than we thought. There are most likely still many other mitochondrial proteins whose effects could prove to be of great importance. Indeed, several of the mitochondrial derived proteins discovered to date have major impacts on metabolism and could act as important therapeutic molecules. Our results lead to rethink the evolution and the selection pressures exerted on the mitochondrial genome and open the door to many future researches which will allow to re-characterize the mitochondrial genome and to have an even deeper understanding of the role of mitochondria in cellular functions.

No description available.

Fusionspeptid-, Redox-, Viruscore- und sonstige Proteine. Alle analysierten Single ORF (SO)-VVOV Rekombinanten vermittelten einen signifikanten Schutz vor einer tödlichen Belastung mit Aujeszky-Virus. Zwei der Rekombinanten (SO 93-, SO 94-VVOV) enthalten ORFs, die für ATI/Fusionspeptid-Proteine kodieren. In SO 19- und SO 70-VVOV sind dagegen für Redoxproteine kodierende ORFs integriert. Weiterführende Untersuchungen zeigten, dass SO 94- und SO 19-VVOV in zwei weiteren Modellsystemen immunstimulatorisch aktiv sind. Im Baculo-Virussystem exprimierte Proteine waren nur in Kombination mit Vaccinia Lister-Virus (VV) wirksam. Dabei zeigten jeweils Virus-Protein-Gemische mit dem geringsten Proteinanteil den stärksten immunstimulatorischen Effekt. Proben in denen VV durch bovines Herpes-Virus-1 ersetzt wurde, sind dagegen nicht wirksam. Dies lässt auf eine Beteiligung VV-spezifischer Faktoren schließen. Übereinstimmend mit diesen Ergebnissen führte eine Frameshift-Mutation in ORF 94r von SO 94mut-VVOV nur zur Abschwächung und nicht zum vollständigen Verlust der immunstimulatorischen Wirkung. Beide in Schizosaccharomyces pombe exprimierten Proteine, sp-ORF19 und sp-ORF94r, induzierten keinen signifikanten Schutz im Aujeszky Maus Modell. Mit der Identifikation einzelner immunstimulatorisch aktiver PPVO-Komponenten ist es erstmals gelungen, den paramunisierenden Effekt von Parapox-Viren einzelnen viralen Genen zu zuordnen. Insbesondere stellen SO 94- und SO 19-VVOV viel versprechende Kandidaten für die prophylaktische bzw. therapeutische Anwendung in verschiedenen Indikationen als auch für weitere Untersuchungen des Wirkmechanismus dar.

Fusionspeptid-, Redox-, Viruscore- und sonstige Proteine. Alle analysierten Single ORF (SO)-VVOV Rekombinanten vermittelten einen signifikanten Schutz vor einer tödlichen Belastung mit Aujeszky-Virus. Zwei der Rekombinanten (SO 93-, SO 94-VVOV) enthalten ORFs, die für ATI/Fusionspeptid-Proteine kodieren. In SO 19- und SO 70-VVOV sind dagegen für Redoxproteine kodierende ORFs integriert. Weiterführende Untersuchungen zeigten, dass SO 94- und SO 19-VVOV in zwei weiteren Modellsystemen immunstimulatorisch aktiv sind. Im Baculo-Virussystem exprimierte Proteine waren nur in Kombination mit Vaccinia Lister-Virus (VV) wirksam. Dabei zeigten jeweils Virus-Protein-Gemische mit dem geringsten Proteinanteil den stärksten immunstimulatorischen Effekt. Proben in denen VV durch bovines Herpes-Virus-1 ersetzt wurde, sind dagegen nicht wirksam. Dies lässt auf eine Beteiligung VV-spezifischer Faktoren schließen. Übereinstimmend mit diesen Ergebnissen führte eine Frameshift-Mutation in ORF 94r von SO 94mut-VVOV nur zur Abschwächung und nicht zum vollständigen Verlust der immunstimulatorischen Wirkung. Beide in Schizosaccharomyces pombe exprimierten Proteine, sp-ORF19 und sp-ORF94r, induzierten keinen signifikanten Schutz im Aujeszky Maus Modell. Mit der Identifikation einzelner immunstimulatorisch aktiver PPVO-Komponenten ist es erstmals gelungen, den paramunisierenden Effekt von Parapox-Viren einzelnen viralen Genen zu zuordnen. Insbesondere stellen SO 94- und SO 19-VVOV viel versprechende Kandidaten für die prophylaktische bzw. therapeutische Anwendung in verschiedenen Indikationen als auch für weitere Untersuchungen des Wirkmechanismus dar.

<p>Epoxides are three-membered cyclic ethers formed in the metabolism of foreign substances and as endogenous metabolites. Epoxide hydrolases (EHs) are enzymes that catalyze the hydrolysis of epoxides to yield the corresponding diols. EHs have been implicated in diverse functions such as detoxification of various toxic epoxides, as well as regulation of signal substance levels.</p><p>The main goal of this thesis was to investigate and characterize the α/β hydrolase fold EH. The first part concerns the identifictaion of an EH in <i>Saccharomyces cerevisiae</i>. The second part involves detailed mechanistic and structural studies of a plant EH from potato, StEH1. </p><p>Despite the important function of EH, no EH has previously been established in <i>S. cerevisiae</i>. By sequence analysis, we have identified a new subclass of EH present in yeast and in a wide range of microorganisms. The <i>S. cerevisiae</i> protein was produced recombinantly and was shown to display low catalytic activity with tested epoxide substrates. </p><p>In plants, EHs are involved in the general defence system, both in the metabolism of the cutin layer and in stress response to pathogens. The catalytic mechanism of recombinantly expressed wild type and mutant potato EH were investigated in detail using the two enantiomers of <i>trans</i>-stilbene oxide (TSO). The proposed catalytic residues of StEH1 were confirmed. StEH1 is slightly enantioselective for the <i>S,S</i>-enantiomer of<i> trans</i>-stilbene oxide. Furthermore, distinct pH dependence of the two enantiomers probably reflects differences in the microscopic rate constants of the substrates. The detailed function of the two catalytic tyrosines was also studied. The behavior of the tyrosine pair resembles that of a bidentate Lewis acid and we conclude that these tyrosines function as Lewis acids rather then proton donors.</p><p>The three dimensional structure of StEH1 was solved, representing the first structure of a plant EH. The structure provided information about the substrate specificity of StEH1.</p>

Epoxides are three-membered cyclic ethers formed in the metabolism of foreign substances and as endogenous metabolites. Epoxide hydrolases (EHs) are enzymes that catalyze the hydrolysis of epoxides to yield the corresponding diols. EHs have been implicated in diverse functions such as detoxification of various toxic epoxides, as well as regulation of signal substance levels. The main goal of this thesis was to investigate and characterize the α/β hydrolase fold EH. The first part concerns the identifictaion of an EH in Saccharomyces cerevisiae. The second part involves detailed mechanistic and structural studies of a plant EH from potato, StEH1. Despite the important function of EH, no EH has previously been established in S. cerevisiae. By sequence analysis, we have identified a new subclass of EH present in yeast and in a wide range of microorganisms. The S. cerevisiae protein was produced recombinantly and was shown to display low catalytic activity with tested epoxide substrates. In plants, EHs are involved in the general defence system, both in the metabolism of the cutin layer and in stress response to pathogens. The catalytic mechanism of recombinantly expressed wild type and mutant potato EH were investigated in detail using the two enantiomers of trans-stilbene oxide (TSO). The proposed catalytic residues of StEH1 were confirmed. StEH1 is slightly enantioselective for the S,S-enantiomer of trans-stilbene oxide. Furthermore, distinct pH dependence of the two enantiomers probably reflects differences in the microscopic rate constants of the substrates. The detailed function of the two catalytic tyrosines was also studied. The behavior of the tyrosine pair resembles that of a bidentate Lewis acid and we conclude that these tyrosines function as Lewis acids rather then proton donors. The three dimensional structure of StEH1 was solved, representing the first structure of a plant EH. The structure provided information about the substrate specificity of StEH1.