Os gêneros Isocheles e Loxopagurus são endêmicos da América e ocorrem em águas tropicais e subtropicais. Isocheles é composto por cinco espécies, duas encontradas no Atlântico oeste (I. sawayai e I. wurdemanni) e três no Pacífico leste (I. pilosus, I. pacificus, e I. aequimanus). Loxopagurus é um gênero monotípico e ocorre apenas na costa sudeste da América do Sul. Estes dois gêneros são morfologicamente semelhantes, tendo como principal diferença a heteroquelia em Loxopagurus em contraste com a isoquelia de Isocheles. Há poucas informações na literatura sobre estes gêneros, e dúvidas quanto a seu status taxonômico foram recentemente ressuscitadas. Dessa forma este estudo visa elucidar as relações filogenéticas de Isocheles e Loxopagurus e avaliar a sua validade taxonômica. Para isso, foram realizadas análises moleculares utilizando os genes 16S rRNA, COI e H3, e análises morfológicas, buscando caracteres que facilitem a identificação destas espécies, bem como a contextualização de ambos os gêneros dentro da Família. Os tipos de I. aequimanus e I. pilosus foram perdidos e erros de identificação entre as espécies de Isocheles são bem comuns. Assim, foram propostos caracteres que delimitam de maneira clara estas espécies, como ornamentação e formato dos quelípodos e o número de dentes presentes no segundo segmento da antena. A proposição de neótipos não foi necessária, uma vez que não há problemas taxonômicos entre estas espécies que não puderam ser resolvidos com base nas descrições originais e desenhos. Um dos lotes de I. wurdemanni do Museum of Comparative Zoology foi identificado pelo presente estudo como o holótipo da espécie. Isocheles e vii Loxopagurus se mostraram gêneros monofiléticos distintos, com base nos 3 marcadores utilizados, e grupos irmãos dentro de Diogenidae. Além disso, constatou-se que as espécies de Isocheles também formam grupos monofiléticos e uma nova espécie foi encontrada, divergindo tanto morfológica quanto molecularmente das demais. Assim, com base na morfologia, na topologia das árvores geradas e nas distâncias genéticas, conclui-se que, não somente os dois gêneros, mas também as espécies englobadas, são unidades taxonômicas distintas válidas. / The genera Isocheles and Loxopagurus are endemic to America and occur in tropical and subtropical waters. There are five species of Isocheles, two of them are found in West Atlantic (I. sawayai and I. wurdemanni) and three are found in East Pacific (I. pilosus, I. pacificus and I. aequimanus). Loxopagurus is a monotypic genus and occurs only in southeast coast of South America. These two genera are morphologically similar, but the main difference between them is the heterochelia of Loxopagurus, while the chelipeds of Isocheles are similar in shape and size. There are few published information about these genera, and some doubts about their taxonomic status arised recently. Thus, this study aims to elucidate the phylogenetic relationship of Isocheles and Loxopagurus and evaluate their taxonomic validity. For that, molecular analyses were performed based on the genes 16S rRNA, COI and H3, as well as morphological analysis, seeking characters that facilitate the identification of these species, and that contextualize both genera in the family. The type specimens of I. aequimanus and I. pilosus were lost and identification errors are common between Isocheles species. Therefore, characters that clearly delimit these species were stated, as the ornamentation and shape of the chelipeds and the number of teeth in the second segment of the antenna. There was no need to designate neotypes, once there are no taxonomic problems between the species that could not be solved based on the original descriptions and illustrations. Also, we identified that one of the lots of I. wurdemanni from the Museum of Comparative Zoology is the holotype of the species. The analysis based on the three ix molecular markers showed that Isocheles and Loxopagurus are two different monophyletic genera, and sister taxa among Diogenidae. We also verified that each one of the species of Isocheles are monophyletic, and found a new species that differs both molecular and morphologically from the others. Thus, based on the morphology, on the tree topology and on genetic divergences, we concluded that not only the two genera, but also the species that they encompass, are valid taxonomic units.

O gênero Pagurus é um táxon heterogêneo de ermitões, com ampla distribuição mundial, descrito há mais de duzentos anos. A sistemática do grupo é complexa com uma longa história de rearranjos taxonômicos. A classificação conta com a inclusão de numerosas novas espécies e separação de algumas inicialmente contidas no táxon estabelecendo-se novos gêneros. Devido à extensa distribuição das espécies que compõem o táxon, foi necessário restringir o objetivo deste estudo. Assim foi avaliado o status taxonômico das espécies que ocorrem nas costas Pacífica e Atlântica da América do Sul, por meio da combinação de análises morfológicas e moleculares das espécies, utilizando dois marcadores genéticos (16S rDNA e Histona 3). As análises taxonômicas mostraram uma alta variabilidade morfológica nas 22 espécies examinadas. As espécies se encaixam perfeitamente em quatro dos onze grupos preestabelecidos dentro de Pagurus. Além disso, foram fornecidos os caracteres morfológicos que definem um desses grupos. Adicionalmente foi incluída uma chave para ajudar na identificação de todas as espécies. As análises independentes dos dados moleculares mostraram resultados contrastantes. O gene mitocondrial foi mais variável e portanto mais informativo, proporcionando uma hipótese mais clara das relações internas entre os membros de Pagurus. Assim, as topologias moleculares resultantes, concordaram em vários aspectos com o reportado nos dados morfológicos das espécies. De modo que, as semelhanças morfológicas foram refletidas na formação dos nós internos. Assim, as análises do gene 16S e H3 mostraram-se concordante com a morfologia, refletindo-se na formação de alguns dos grupos previamente propostos. Como ponto importante, ressalta-se a separação das espécies que compõe o grupo provenzanoi como um táxon diferenciado dentro de Pagurus. Ao mesmo tempo, as análises com o gene H3 mostraram a espécie Propagurus gaudichaudii inserida dentro de Pagurus, questionando a validade taxonômica de Propagurus. Como só foi incluída uma das cinco espécies do gênero, é claramente necessário a inclusão de outras espécies contidas neste táxon, bem como alguns outros genes uma revisão das espécies contidas neste táxon, junto com análises de outros genes, a fim de resolver definitivamente o status taxonômico de Propagurus. / The genus Pagurus is a heterogeneous taxon of hermit crabs, with worldwide distribution which was described more than two hundred years ago. Systematics of the group is complex and with a long history of taxonomic rearrangements. Thus current classification accounts for inclusion of many new species and splitting-off of some of the original species in order to establish new genera. Because of the extensive distribution of the species that conform this taxon it was necessary to restrict the aim of this study. Thus I evaluated the taxonomic status of the species found in the Pacific and Atlantic coasts of South America using a combination of both morphologic and molecular analyses (16S rDNA and Histona 3). The taxonomic analysis showed high morphological variation among all the 21 examined species. The analyzed species seemed to perfectly fit four out of the eleven morphologically pre-established groups of Pagurus. Furthermore I provide morphological characters that define one of these groups. Additionally, I included a key to aid in the identification of all the target species. Independent analysis of the molecular data showed contrasting results. The mitochondrial gene was the most variable and thus the most informative, providing a clearer hypothesis of the internal relationships among members of Pagurus. Therefore, both 16S and H3 analyses were in general agreement with the morphology. Thus, the resultant molecularly based topologies reflected some of the groups previously proposed. It is important to point out that all the included species belonging to the provenzanoi group were clustered together separated from other clades within Pagurus. At the same time, the analysis with the gen H3 showed Propagurus gaudichaudii clustered within Pagurus, thus questioning the taxonomic validity of Propagurus. As I only included one of the five species of the genus, it is obviously necessary to include other the species contained in this taxon, as well as some other genes in order to definitely solve the taxonomic status of Propagurus.

Neutrophils are major cell type of innate immunity, that can eliminate pathogens by different mechanisms. One of these mechanisms is called NETosis, which leads to release of decondensed chromatin and citrullinated histone proteins. Citrullination is post-translational modification catalysed by peptidylarginine deiminase (PAD) and causing transformation of possitively charged arginin to neutral citrullin and can change expression of cytokine genes. Concetrations of pro-inflammatory cytokines (IL-8, TNF, IL-1) were measured after activation of PAD4 and induction of citrullination. Calcium ionophore was used to induce citrullinaton, Cl-amidine and TDFA were used as inhibitors. Production of cytokines was assessed by ELISA on protein level and by qPCR on mRNA level. It was found that induction of citrullination led to increased concentrations of IL-8 and IL-1. Elevated gene expression of IL-8 was confirmed on mRNA level. Both inhibitors were able to decrease level of histone H3 citrullination and IL-8 and IL-1 concentrations. Expression of TNF was not detected on protein and mRNA level.

Les gliomes infiltrants du tronc cérébral (DIPG) sont les gliomes pédiatriques les plus agressifs. Leur localisation dans le pont de Varole et leur caractère infiltrant les rendent inopérables. Le manque de matériel biologique et l’absence de modèles pertinents ont longtemps freiné le développement de nouvelles thérapies. Le séquençage du génome entier de ces tumeurs a identifié la substitution K27M des gènes d’histone HIST1H3B/C et H3F3A comme mutation somatique initiatrice des DIPG. Bien que partageant la mutation sur le même résidu, notre laboratoire a montré que les tumeurs mutées dans le gène codant H3.1 ou H3.3 définissent des sous-groupes de DIPG avec des programmes oncogéniques, propriétés d’invasion et survies différents.Dans ce contexte, le premier objectif de ma thèse a consisté à développer des modèles in vitro et in vivo pertinents de la maladie. Pour ce faire, nous avons dissocié des biopsies fraiches de DIPG naïves de tout traitement et avons, de manière systématique, mis en culture une moitié de la suspension dans du milieu sans sérum pour permettre l’expansion de cellules souches de gliomes (GSC). En parallèle, l’autre moitié a été directement injectée par stéréotaxie dans le cerveau de souris immunodéprimées, pour obtenir des modèles de xénogreffes dérivés de patients (PDX). De plus, nous avons obtenu un second type de modèle in vivo de DIPG à partir des cellules tumorales en culture greffées dans le tronc par stéréotaxie (CDX). Au préalable, les GSC ont été modifiées pour exprimer de manière stable la protéine fluorescente mKate2 et l’enzyme Luciférase, permettant le suivi longitudinal de la croissance tumorale. Nous avons pu générer des GSCs issues de 26 patients différents mutées dans le gène codant H3.1 ou H3.3 et qui maintiennent leurs propriétés de cellules souches in vitro. De plus, nous avons généré 8 modèles PDX et 6 modèles CDX développant des tumeurs infiltrantes et présentant les mêmes symptômes neurologiques que les patients. Conformément aux données cliniques, nos modèles in vivo mutés H3.3 ont développé des tumeurs infiltrant le parenchyme plus rapidement que les modèles H3.1. En somme, ces résultats indiquent que nous avons réussi à générer des modèles de DIPG à partir d’échantillons naïfs de tout traitement, et qui récapitulent les aspects cliniques et moléculaires de ce cancer.Nous avons ensuite profité des modèles GSCs pour étudier les mécanismes moléculaires qui conduisent au phénotype invasif. Dans un premier temps, nous avons montré que l’expression de CXCR4, un récepteur de chimiokines, est corrélée à la capacité migratoire des GSC in vitro. Cependant, l’inhibition de CXCR4 par un antagoniste n’a montré d’efficacité que dans une partie des GSC testées, suggérant l’implication d’autres voies de signalisations dans leur motilité. Aussi, dans le but d’identifier de manière non biaisée de nouveaux mécanismes moléculaires impliqués dans la migration des GSC, nous avons analysé le profil transcriptomique de 13 GSC présentant une mutation K27M dans les gènes codant H3.1 ou H3.3 par séquençage de l’ARN. Nous avons ensuite sélectionné les gènes dont le niveau d’expression était corrélé à la vitesse de migration des différentes GSCs. Ainsi, nous avons découvert plusieurs candidats, parmi lesquels CXCL10, le ligand du récepteur aux chimiokines CXCR3, et SPSB4, une ubiquitine-ligase, qui sont surexprimés dans les tumeurs migrant rapidement. L’implication de ces 2 gènes dans la migration sera étudiée in vitro et in vivo dans des expériences de perte de fonction en utilisant des constructions exprimant des shARN inductibles.Au total, ce travail a permis le développement de modèles pertinents de DIPG qui récapitulent les caractéristiques de la tumeur dont ils dérivent et qui reflètent l’hétérogénéité interindividuelle observée chez les patients. De plus, cette étude a permis la découverte de nouveaux mécanismes moléculaires qui favorisent la migration de ces cellules tumorales très particulières. / Diffuse Instrincic Pontine Gliomas (DIPG) are the most aggressive form of pediatric gliomas. They are infiltrative tumors originating from the brainstem. The lack of biological material and the absence of relevant models have for long hampered the development of new therapeutics. Whole genome sequencing of tumor DNA identified a K27M substitution in the histone genes HIST1H3B/C and H3F3A as a recurrent somatic driver mutation in DIPG. Despite sharing the mutation on the same residue, our lab demonstrated that H3.1- and H3.3-mutated tumors define distinct subgroups of DIPG with different oncogenic programs, overall survival or invasion properties.Thus, the first goal of my PhD project consisted in developing relevant in vitro and in vivo models from treatment-naive primary specimen of DIPG that would mimic the disease. To do so, our strategy has been to dissociate fresh biopsies upon reception from the operating room and systematically place one-half of the cell suspension in serum-free culture to allow the expansion of Glioma Stem-like cells (GSCs) while the other half is directly stereotactically injected into the brain of Nude mice to develop Patient-derived xenograft models (PDX). We also obtained a second type of in vivo DIPG models, Cell-derived Xenograft models (CDX), after stereotactic xenografting of GSCs. Cells were modified beforehand to stably express a fluorescent protein as well as the Firefly luciferase enzyme, thus allowing the longitudinal record of tumor growth. We succeeded in generating GSCs from 26 different patients harboring a K27M mutation either in H3.1 or H3.3, that maintained stemness properties in culture. In addition, we generated 8 PDX and 6 CDX models that developed infiltrative tumors with maintenance of DIPG hallmarks. All tumor-bearing mice displayed neurological disorders like impaired walking and/or hemi-paralysis, as observed in patients. In accordance with the clinical data, our H3.3-K27M in vivo models developed tumors faster than H3.1-K27M, as well as they infiltrate the parenchyma faster and deeper than H3.1-K27M models. Taken together, these results showed that we succeeded in generating models of DIPG from treatment-naïve samples that mimic the clinical and molecular aspects of the disease.We then took advantage of such GSCs models to decipher the molecular mechanisms driving the invasive phenotype of patients with DIPG. First, we showed that CXCR4 expression, a chemokine receptor, is correlated with the in vitro migration capacity of our cellular models of DIPG. However, inhibition using CXCR4-antagonist CXCR4 was only efficient in a subset of our models, suggesting that other signaling pathways could be involved in their motility. To further identify the molecular mechanisms involved in migration of GSCs in vitro in an unbiased manner, we profiled the transcriptome of 13 GSCs harboring both H3.1- and H3.3-K27M genotype by RNA-sequencing. We then selected genes which expression levels correlated with the migration speed of the different GSCs. We uncovered several candidates among which CXCL10, the ligand of the chemokine receptor CXCR3, and SPSB4, an ubiquitin-ligase, up-regulated in highly migrating tumor cells. The involvement of these 2 genes in migration will be investigated in vitro and in vivo in loss of function experiments using doxycycline-inducible shRNA expressing constructs.Altogether, this work allowed for the development of pertinent DIPG models that resemble the primary tumors they derive from and also reflect the inter-individual heterogeneity observed in patients. Additionally, we gained new insights on the molecular mechanisms promoting cell migration in these particular cells.

Protein Prp45, the yeast ortholog of the human transcription coregulator SNW1/SKIP, has been previously associated only with the regulation of pre-mRNA splicing. However, our laboratory found that protein Prp45 genetically interacts not only with the proteins involved in pre-mRNA splicing, but also with factors important for transcription elongation and with chromatin modifying enzymes. Our data and the information about the human ortholog SNW1/SKIP suggest that Prp45 could serve as a regulator coupling splicing, transcription and chromatin state in S. cerevisiae. The main aim of this diploma thesis was to find out whether the protein Prp45, which is essential for cotranscriptional assembly of the spliceosome, affects posttranslational modifications of chromatin on transcribed genes. Using chromatin immunoprecipitation, the influence of prp45(1-169) mutation on trimethylation of histone H3 at lysine 4 and acetylation of histone H3 at lysines 9, 14 and 18 on transcriptionally active genes was not confirmed. The other aim was to analyse the behavior of cells synchronized by α-factor by using flow cytometry. According to our results, prp45(1-169) mutation leads to the prolongation of the cell cycle. For the purpose of monitoring the dynamics of nucleosomes in S. cerevisiae strains, the system of...

Relations among transcription, pre-mRNA processing and chromatin modifications are only partially understood. The human protein SNW1/SKIP belongs to factors which couple these processes. The protein plays role in pre-mRNA splicing and transcription on the level of both initiation and elongation. According to the hypothesis of K. Jones laboratory, it physically and functionally interacts with positive transcription elongation factor b during transcription elongation and influences methylation of histone H3 on lysine 4, a modification characteristic for active transcription (Bres et al., Genes Dev. 19:1211-26, 2005, Bres et al., Mol Cell. 36:75-87, 2009). The yeast ortholog of SNW1/SKIP, Prp45, was until now reported only in connection with splicing regulation. However, unpublished results from our Laboratory and others showed that it is employed in transcription elongation as well. The aim of the diploma project was to search for the relations between Prp45 and the factors regulating transcription. It was confirmed that the mutation prp45(1 169) results in the delay of PHO5 and PHO84 expression during transcriptional induction. Next, we discovered new genetic interactions between PRP45 and several genes encoding the effectors of chromatin modifications. How Prp45 influences the expression of PHO5 and PHO84...

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Normal cellular behaviour in multicellular organisms is achieved by tight control of signaling pathway networks. The mitogen-activated protein kinase (MAPK) signaling cascade is one of these signaling networks, that when deregulated can lead to cellular transformation. Activation of the RAS-RAF-MEK-MAPK (ERK) signal transduction pathway or the SAPK2/p38 pathway results in the activation of mitogen- and stress-activated protein kinases 1 and 2 (MSK1/2). Subsequently, MSKs go on to phosphorylate histone H3 at Ser10 and Ser28.Here, we demonstrate that the activities of ERK and MSK1, but not p38, are elevated in Hras-transformed cells (Ciras-3) relative to these activities in the parental 10T1⁄2 cells. Analyses of the subcellular distribution of MSK1 showed that the H3 kinase was similarly distributed in Ciras-3 and 10T1/2 cells, with most MSK1 being present in the nucleus. In contrast to many other chromatin modifying enzymes, MSK1 was loosely bound in the nucleus and was not a component of the nuclear matrix. Our results provide evidence that oncogene-mediated activation of the RAS-MAPK signal transduction pathway elevates the activity of MSK1, resulting in the increased steady-state levels of phosphorylated H3, which may contribute to the chromatin decondensation and aberrant gene expression observed in oncogene-transformed cells. Furthermore, upon activation of the ERK and p38 MAPK pathways, the MSK1/2- mediated nucleosomal response, including H3 phosphorylation at serine 28 or 10, is coupled with the induction of immediate-early gene transcription. The outcome of this response, varying with the stimuli and cellular contexts, ranges from neoplastic transformation to neuronal synaptic plasticity. Here, we used sequential co-immunoprecipitation assays and chromatin immunoprecipitation (ChIP) assays on mouse fibroblast 10T1/2, Ciras-3 and MSK1 knockdown 10T1/2 cells to show that H3 serine 28 and 10 phosphorylation leads to promoter remodeling. MSK1, in complexes with phospho-serine adaptor 14-3-3 proteins and BRG1 (the ATPase subunit of the SWI/SNF remodeler) is recruited to the promoter of target genes by transcription factors such as ELK-1 or NFκB. Following MSK1-mediated H3 phosphorylation, BRG1 associates with the promoter of target genes via 14-3-3 proteins, which act as scaffolds. The recruited SWI/SNF remodels nucleosomes at the promoter of immediate-early genes enabling the binding of transcription factors like JUN and the onset of transcription. Since RAS-MAPK activated MSKs mediate H3 phosphorylation that is required for expression of various immediate-early gene products involved in cellular transformation, inhibition of MSK activity may be a therapeutic target that could be exploited in cancers with upregulated RAS-MAPK signaling.

The RAS-mitogen-activated protein kinase (MAPK) pathway is an essential signaling mechanism that regulates cellular processes and culminates in the activation of specific gene expression programs. Alterations in the RAS-MAPK signaling cascade can modify epigenetic programs and confer advantages in cell growth and transformation. In fact, deregulation of the cascade is a key event in tumour development with 30% of human cancers harbouring RAS mutations. In breast and pancreatic epithelial cancers, characterization of an aberrant RAS-MAPK pathway has focused on upstream mediators such as receptors and oncogenic RAS molecules but the impact of downstream targets remain poorly defined. Stimulation of the RAS-RAF-MEK-MAPK pathway leads to activation of mitogen- and stress-activated protein kinases 1 and 2 (MSK1/2) which are responsible for the phosphorylation of histone H3 on S10 and S28. We postulate that deregulation of the RAS-MAPK pathway produced by constitutive activation and/ or over-expression of upstream components or mitogen stimulation consequently leads to enhanced MSK1 activity and elevated histone H3 phosphorylation levels. We further hypothesize that MSK1-mediated H3 phosphorylation is critical for immediate early gene (IEG) expression, Ras-driven transformation and is associated with regulatory regions upon gene transcription. In mouse fibroblasts, we present evidence for the critical involvement of MSK1 and H3 phosphorylation as mediators that bridge the aberrant signals driven by the RAS-MAPK pathway with nucleosomal modifications, chromatin remodeling, IEG expression and malignant transformation. We then examined if activation of RAS-MAPK signaling in breast cancer cells elicits similar molecular events. We demonstrate that the RAS-MAPK pathway is induced and enhances the association of MSK1 and H3 phosphorylation on the IEG Trefoil Factor 1 resulting in transcriptional activation. We further observed that mutated K-RAS expression did not correlate with genomic instability or altered signaling in pancreatic cancer cell lines while overexpressed HER2 and EGFR breast cancer cell lines generally exhibit upregulated ERK1/2 and H3 phosphorylation levels. Taken together, our studies contribute to the further understanding of MSK-mediated transcriptional activation in response to RAS-MAPK signaling in oncogene-transformed and cancer cell lines. Inhibition of MSK activity may be an unexplored avenue for combination cancer therapy with abnormal RAS-MAPK signaling pathways.

Chromatin is the combination of DNA and proteins in the nucleus that is used to aid in the compaction of DNA. Histones are a group of proteins used to condense DNA by forming a complex (nucleosome) around which DNA wraps around; there are two of each type of histone in a nucleosome: H2A, H2B, H3 and H4. Once the DNA is wrapped around the histones, the genome is further compacted. A shortened, "clipped" version of histone H3 has been found in some organisms including yeasts, flies, mammalian stem cells, and the ciliated protozoan, Tetrahymena thermophila. In each organism, clipping occurs at a different site on the N-terminus, usually before an alanine residue. Clipping is important as it may affect other epigenetic modifications and gene regulation in cell differentiation, but the regulation of this histone proteolysis has remained largely unstudied. In Tetrahymena thermophila, approximately half of the histone H3 molecules are clipped between residues 6 and 7 on histone H3, solely in the transcriptionally silent micronucleus. The histones in the micronuclei are deacetylated, while histones in the macronuclei can be acetylated or deacetylated. It is hypothesized that the post-translational acetylation modification to the histone tails may inhibit histone H3 clipping. Immunoblot analyses were carried out with acetylated and deacetylated micronuclei, demonstrating an increase of clipping when acetylated. Additionally, mutations were created at lysine 9 upstream of the clip site on the histone H3 tails to mimic acetylation and deacetylation to study whether the modification of that site has a regulatory effect.