Tandem mass spectrometry (MS/MS) is the dominant approach for large-scale peptide sequencing in high-throughput proteomic profiling studies. The computational analysis of MS/MS spectra involves the identification of peptides from experimental spectra, especially those with post-translational modifications (PTMs), as well as the inference of protein composition based on the putative identified peptides. In this thesis, we tackled two major challenges associated with an MS/MS analysis: 1) the refinement of PTM predictions from MS/MS spectra and 2) the inference of protein composition based on peptide predictions. We proposed two PTM prediction refinement algorithms, PTMClust and its Bayesian nonparametric extension \emph{i}PTMClust, and a protein identification algorithm, pro-HAP, that is based on a novel two-layer hierarchical clustering approach that leverages prior knowledge about protein function. Individually, we show that our two PTM refinement algorithms outperform the state-of-the-art algorithms and our protein identification algorithm performs at par with the state of the art. Collectively, as a demonstration of our end-to-end MS/MS computational analysis of a human chromatin protein complex study, we show that our analysis pipeline can find high confidence putative novel protein complex members. Moreover, it can provide valuable insights into the formation and regulation of protein complexes by detailing the specificity of different PTMs for the members in each complex.

Tandem mass spectrometry (MS/MS) is the dominant approach for large-scale peptide sequencing in high-throughput proteomic profiling studies. The computational analysis of MS/MS spectra involves the identification of peptides from experimental spectra, especially those with post-translational modifications (PTMs), as well as the inference of protein composition based on the putative identified peptides. In this thesis, we tackled two major challenges associated with an MS/MS analysis: 1) the refinement of PTM predictions from MS/MS spectra and 2) the inference of protein composition based on peptide predictions. We proposed two PTM prediction refinement algorithms, PTMClust and its Bayesian nonparametric extension \emph{i}PTMClust, and a protein identification algorithm, pro-HAP, that is based on a novel two-layer hierarchical clustering approach that leverages prior knowledge about protein function. Individually, we show that our two PTM refinement algorithms outperform the state-of-the-art algorithms and our protein identification algorithm performs at par with the state of the art. Collectively, as a demonstration of our end-to-end MS/MS computational analysis of a human chromatin protein complex study, we show that our analysis pipeline can find high confidence putative novel protein complex members. Moreover, it can provide valuable insights into the formation and regulation of protein complexes by detailing the specificity of different PTMs for the members in each complex.

Orientador: Jorg Kobarg / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Biologia / Made available in DSpace on 2018-08-07T08:03:42Z (GMT). No. of bitstreams: 1 Passos_DarioOliveirados_D.pdf: 4831709 bytes, checksum: 0aa3e031d4e82dc447636416b68401e8 (MD5) Previous issue date: 2006 / Resumo: A proteína Ki-1/57 que é encontrada tanto no núcleo quanto no citoplasma está associada com atividade de proteína quinase serina/treonina e é fosforilada nestes resíduos após ativação celular. Neste trabalho verificamos que Ki-1/57 interage com a proteína Chromatin-Helicase-DNA-binding domain 3 (CHD3) e com a proteína adaptadora/sinalizadora RACK1 no núcleo. Pelo sistema do duplo híbrido de levedura (SDHL) a proteína arginina metiltransferase 1 (PRMT1) foi selecionada como outra proteína de interação. A PRMT1 integra uma família representada por nove enzimas humanas que catalisam reações de metilação em resíduos de arginina. Em seguida, usando agora a PRMT1 como isca - no SDHL - identificamos as proteínas Ki-1/57 e hnRNPQ, juntamente com outras 13. A maioria delas contêm motivos ¿RGG-box¿ em suas seqüências de aminoácidos, que são conhecidos alvos para metilação. Posteriormente verificamos que Ki-1/57 e seu provável parálogo CGI-55 conservam dois motivos ¿RGG/RXR-box¿ e que são substratos in vitro para a metilação de argininas pela PRMT1. Estudos de mapeamento mostraram que todos os fragmentos contendo o motivo ¿RGG/RXR-box¿ interagem com a PRMT1 e são alvos à metilação in vitro. Ki-1/57 endógena, imunoprecipitada de células L540, mostrou ser metilada in vivo, além de ser um alvo a metilação pela PRMT1 in vitro, somente quando as células são previamente tratadas com o inibidor da metilação Adox. Tratamento das células Hela com o inibidor da metilação (Adox) causa desaparecimento da imuno-marcação citoplasmática de Ki-1/57 e relativa redistribuição do parálogo CGI-55 para o citosol. Assim, pode ser especulado que a metilação destas proteínas deve ser um evento importante para suas localizações subcelulares e conseqüentemente para suas funções. Em resumo, nossos dados sugerem que o SDHL é um método efetivo na identificação de novos substratos celulares para a PRMT1 e poderia ser estendido para a identificação e caracterização de novos substratos para os outros integrantes da família das PRMTs humanas / Abstract: The protein Ki-1/57 that is found both in the cytoplasm and nucleus is associated with serine/threonine protein kinase activity and gets phosphorylated on serine and threonine residues upon cellular activation. We demonstrated that Ki-1/57 interacts with the Chromatin-Helicase-DNA-binding domain protein 3 (CHD3) and with the adaptor/signaling protein RACK1 in the nucleus. By utilizing the yeast two-hybrid system (YTHS), we were further able to find the protein arginine-methylatranseferase-1 (PRMT1) as another interacting protein. PRMT1 is a member of the family constituted by 9 human enzymes that catalyze methylation reactions on arginine residues. Afterwards, by using PRMT1 as bait in the YTHS we identified both Ki-1/57 and NSAP1 as interacting proteins, along with 13 other proteins. The majority of them present RGG-box clusters in their amino acid sequences, which are known to be targets for arginine methylation. We further found that Ki-1/57 and its putative paralogue CGI-55 have two RGG/RXR-box clusters conserved between them and that they are substrates for arginine-methylation by PRMT1 in vitro. In mapping studies, we observed that all Ki-1/57 protein fragments containing the RGG/RXRbox clusters interact with PRMT1 and are targets for methylation in vitro. Endogenous cellular Ki-1/57 seems to be methylated in vivo and is a target for methylation by PRMT1 in vitro, only when cells have been previously treated with the methylation inhibitor Adox. Treatment of Hela cells with the inhibitor of methylation (Adox) causes the disappearance of the immuno-staining of Ki-1/57 in the cytoplasm and a relative redistribution of the paralogue CGI-55 to the cytosol. It can therefore be speculated that the methylation of these proteins is important for their sub-cellular localization and in consequence for their function. In summary our data suggest that the YTHS is an effective method for the identification of novel cellular PRMT substrates and could be extended for the identification and characterization of novel substrates to the other components of the human PRMT1 family / Doutorado / Bioquimica / Doutor em Biologia Funcional e Molecular

Ces vingt dernières années, les réactions d'insertion d'atomes ou de fragments moléculaires dans des liaisons C-H peu réactives ont fait l'objet de nombreuses études sans que les mécanismes de ces réactions aient pu être établis. Les enzymes de la superfamille « Radical-SAM » catalysent l'activation de leur substrat en utilisant un centre [4Fe-4S] et le co-substrat S-adénosylméthionine (SAM). Les enzymes d'insertion radicalaire constituent un sous-groupe de cette famille et contiennent un second centre fer-soufre impliqué, lui, dans l'activation du deuxième substrat rendant ainsi possible la réaction d'insertion par couplage radicalaire. Le travail présenté dans cette thèse concerne deux de ces enzymes, la première, RimO, est une méthylthiotransférase (MTTase) qui catalyse l'insertion d'un groupement thiométhyle en beta du résidu D89 de la protéine ribosomale S12 (β-ms-D89-S12). La seconde TYW1 ou 4-demethylwyosine synthase catalyse l'insertion d'un groupement acétyle dérivé du pyruvate dans une liaison C-H d'un groupement N-CH3 appartenant à une guanine spécifique de certains ARNt eucaryotes. Cette réaction d'insertion est suivie d'une cyclisation conduisant en plusieurs étapes à la wybutosine (yW), une base tricyclique importante pour la fidélité traductionnelle de la cellule. Dans ce travail il a été montré que les deux centres de cette famille d'enzyme coopèrent pour ces réactions et contrôlent l'utilisation des différents acteurs par des mécanismes redox originaux. / Over the last twenty years, the insertion reactions of atoms or molecular fragments into poorly reactive C-H bonds have been actively investigated but the details of their mechanisms remain largely unknown. Enzymes belonging to the "Radical-SAM" superfamily catalyze the activation of their substrate using a [4Fe-4S] in conjunction with the co-substrate S-adenosylmethionine (SAM). Radical insertion enzymes are a subgroup of this family and contain a second iron-sulfur cluster involved in the activation of the second substrate allowing the insertion reaction by radical coupling to take place. The work presented in this thesis is focusing on two enzymes, the first one, RimO is a methylthiotransferase (MTTase) that catalyzes the insertion of a thiomethyl group on the beta position of D89 residue of the ribosomal protein S12 (β-ms-D89-S12). The second one, TYW1, or 4-demethylwyosine synthase, catalyzes the insertion of the acetyl moiety of pyruvate into a C-H bond of a N-methyl group of a guanine derivative in some eukaryotic and archeal tRNAs. This insertion reaction leads to the formation of a tricyclic ring and through several steps to wybutosine (yW), a hypermodified nucleotide important for the translational fidelity of the cell. In this work we demonstrate that these radical inserting enzymes utilize the two iron-sulfur clusters to cooperate and that they control the different partners of the reaction by original redox mechanisms.

Histone post-translational modifications (PTMs), as the histone code assumes, are related with regulation of gene transcription, an important mechanism of cells in the differentiation process. Many PTMs are simultaneously present in histone proteins, and changes in the PTM stoichiometric ratios can have several effects, like changes in the chromatin structure leading to a transcriptionally active or repressive state. Significant progresses were made to map variations of histone PTMs by mass spectrometry (MS), and although many protocols were developed there are still some drawbacks. Incomplete and side reactions were identified, which can directly affect the quantification of histone PTMs, because both (incomplete and side reactions) can be misinterpreted as endogenous histone post translational modifications. Therefore, a protocol for derivatization of histones with no noticeable undesired reactions (<10%) was required. In this thesis a new chemical modification methodology is presented, which allows the improvement of sequence coverage by acylation with propionic anhydride of lysine residues and N-terminal (free ε- and α- amino groups) and trypsin digestion. more than 95% of complete reaction was achieved with the new derivatization methodology. This strategy (chemical derivatization of histones), in combination with bottom-up MS approach, allows the quantification of lysine methylation (Kme) and acetylation (Kac) in histones from Saccharomyces cerevisiae (S.cerevisiae), mouse embryonic stem cells (mESCs) and human cell lines. The results showed histone H3 PTM pattern as the most variable profile regarding histone Kme and Kac across the three different organisms and experimental conditions. Therefore, it was concluded that quantification of H3 PTM pattern can be used to examine changes in chromatin states when cells are subjected to any kind of perturbation.

Indiana University-Purdue University Indianapolis (IUPUI) / Papillomaviruses (PVs) are small, double-stranded DNA viruses. Hundreds of species have evolved to replicate in mammals, birds, and reptiles. Approximately two hundred species are estimated to infect humans alone, and these human papillomaviruses (HPVs) cause diseases ranging from benign warts to anogenital and oropharyngeal cancers. While vaccination is effective at preventing the majority of these infections and their disease outcomes, there are no successful treatments for existing infections; thus, exploration of novel therapeutic targets is warranted. PVs control expression and function of their gene products through alternative splicing, alternate start codons, and post-translational modification (PTM). The viral E2 protein regulates transcription, replication, and genome maintenance in infected cells, and PTMs have been demonstrated for E2 proteins from multiple papillomavirus types. Serine phosphorylation events were reported to influence E2 stability, and our laboratory was the first to describe in vitro acetylation events with implications for E2 transcription function. Here we report confirmation of these acetylation events in vivo and additional data elucidating the role of these PTMs in viral transcription. Moreover, we present a novel phosphorylation site for bovine papillomavirus type 1 (BPV-1) E2 at tyrosine 102 (Y102). Using phospho-deficient and phospho-mimetic point mutants, we found that this site influences E2-mediated transcription and replication, and we hypothesize that phosphorylation at Y102 regulates these activities by interrupting the association of E2 with its binding partners. We also report interaction of BPV-1 E2 and HPV-31 E2 with different receptor tyrosine kinases (TKs), most notably members of the fibroblast growth factor receptor family. We hypothesize that Y102 phosphorylation by these receptors occurs early in infection to limit viral replication and gene expression. Further studies will cement the role of RTKs in PV biology and could reveal novel therapeutic strategies.

Indiana University-Purdue University Indianapolis (IUPUI) / Histones are highly basic proteins which when digested by trypsin are hard to analyze using mass spectrometry. Because histones are basic nuclear proteins, a nuclei prep followed by acid extraction is the best purification strategy to increase overall abundance of purified histones. Blocking the lysine residues and cleaving with trypsin is a useful technique to increase detection of histone peptides using MudPIT. In particular, carbamylation and propionylation are the best two methods to block lysine residues. Using both propionylation and carbamylation along with no treatment has been shown to increase the identification of unmodified and modified histone peptides when coupled with MudPIT analysis.

No description available.

No description available.

No description available.