No description available.

<p>These studies are based on the isolation of ë replication defective mutants that had acquired multiple point mutations within ë replication initiation genes O and P in a cryptic prophage (Hayes et al., 1998). Each mutant cell arose after shifting wild type cells with a cI[Ts] cryptic ë prophage deleted for int-kil, and from ren into E. coli, from 30oC to 42oC. Derepression of the trapped cryptic prophage kills the host cells (designated as RK+). Rare colony forming units survive and were designated as RK- mutants. This led to a hypothesis that ë replication-triggered cell stress provokes mutator activity, i.e., increases the frequency of replication errors within the simultaneously replicating chromosome of the host E. coli cells. We tested this hypothesis by asking three questions: (1) Do unselected, untargeted (with no link to ë fragment) auxotrophic mutations appear within the RK- mutant population selected from RK+ culture cells? (2) Is replication initiation from the cryptic ë fragment, or, alternatively, just expression of one or more ë genes required for the appearance of the unselected auxotrophic mutations? (3) Do E. coli functions participate in the appearance of unselected auxotrophic mutations within the RK- mutant population? Our results indicate that auxotrophic mutations unlinked to the ë fragment appeared at high frequency within RK- mutants. RK- auxotrophs arising on rich medium were identified by screening the survivor clones for growth on minimal medium. The appearance of RK- auxotrophic colonies at high frequency (>1 per 100 RK- mutants) leads us to conclude that auxotrophic mutations arise during the independent selection for RK- mutants. Conditions that inhibited ë fragment induction fully suppressed the mutator phenotype. Mutation of host dnaB such that the helicase does not support replication initiation from the induced ë fragment completely suppressed host cell killing, but not the appearance of auxotrophic mutations. We asked if E. coli error-prone polymerases IV and V, or gene functions regulated as part of the host SOS response contributed to the provoked mutator phenotype and observed no close correlation. We demonstrated that the RK+ starting cells did not have a distinct intrinsic mutator activity in several ways, including moving the cryptic ë fragment to different E. coli host cells, blocking ë fragment induction by the addition of a cI+ plasmid to eliminate ë gene expression at high temperatures, and independent assays for spontaneous rifampicin resistance. We found that the induced mutator phenotype associated with the appearance of untargeted auxotrophs was linked to the expression of lambda gene P, and did not require replication initiation from the cryptic ë prophage. We also found that the mutator phenotype of the induced cryptic ë fragment increased the frequency of rifampicin resistant colonies among the RK- mutant population. </p>

<p>These studies are based on the isolation of ë replication defective mutants that had acquired multiple point mutations within ë replication initiation genes O and P in a cryptic prophage (Hayes et al., 1998). Each mutant cell arose after shifting wild type cells with a cI[Ts] cryptic ë prophage deleted for int-kil, and from ren into E. coli, from 30oC to 42oC. Derepression of the trapped cryptic prophage kills the host cells (designated as RK+). Rare colony forming units survive and were designated as RK- mutants. This led to a hypothesis that ë replication-triggered cell stress provokes mutator activity, i.e., increases the frequency of replication errors within the simultaneously replicating chromosome of the host E. coli cells. We tested this hypothesis by asking three questions: (1) Do unselected, untargeted (with no link to ë fragment) auxotrophic mutations appear within the RK- mutant population selected from RK+ culture cells? (2) Is replication initiation from the cryptic ë fragment, or, alternatively, just expression of one or more ë genes required for the appearance of the unselected auxotrophic mutations? (3) Do E. coli functions participate in the appearance of unselected auxotrophic mutations within the RK- mutant population? Our results indicate that auxotrophic mutations unlinked to the ë fragment appeared at high frequency within RK- mutants. RK- auxotrophs arising on rich medium were identified by screening the survivor clones for growth on minimal medium. The appearance of RK- auxotrophic colonies at high frequency (>1 per 100 RK- mutants) leads us to conclude that auxotrophic mutations arise during the independent selection for RK- mutants. Conditions that inhibited ë fragment induction fully suppressed the mutator phenotype. Mutation of host dnaB such that the helicase does not support replication initiation from the induced ë fragment completely suppressed host cell killing, but not the appearance of auxotrophic mutations. We asked if E. coli error-prone polymerases IV and V, or gene functions regulated as part of the host SOS response contributed to the provoked mutator phenotype and observed no close correlation. We demonstrated that the RK+ starting cells did not have a distinct intrinsic mutator activity in several ways, including moving the cryptic ë fragment to different E. coli host cells, blocking ë fragment induction by the addition of a cI+ plasmid to eliminate ë gene expression at high temperatures, and independent assays for spontaneous rifampicin resistance. We found that the induced mutator phenotype associated with the appearance of untargeted auxotrophs was linked to the expression of lambda gene P, and did not require replication initiation from the cryptic ë prophage. We also found that the mutator phenotype of the induced cryptic ë fragment increased the frequency of rifampicin resistant colonies among the RK- mutant population. </p>

La réplication de l’ADN est un processus extrêmement complexe, impliquant des milliers de fourches de réplication progressant le long des chromosomes. Ces fourches sont fréquemment ralenties ou arrêtées par différents obstacles, tels que des structures secondaires de l’ADN, des protéines agissant sur la chromatine ou encore un manque de nucléotides. Ce ralentissement, qualifié de stress réplicatif, joue un rôle central dans le développement tumoral. Des processus complexes, qui ne sont pas encore totalement connus, sont mis en place pour répondre à ce stress. Certaines nucléases, comme MRE11 et DNA2, dégradent l’ADN néorépliqué au niveau des fourches bloquées, ce qui permet le redémarrage des réplisomes.La voie interféron est un mécanisme de défense contre les agents pathogènes qui détecte la présence d’acides nucléiques étrangers dans le cytoplasme et active la réponse immunitaire innée. Des fragments d’ADN issus du métabolisme de l’ADN génomique (réparation, rétrotransposition) peuvent diffuser dans le cytoplasme et activer cette voie. Une manifestation pathologique de ce processus est le syndrome d’Aicardi-Goutières, une maladie rare caractérisée par une inflammation chronique générant des problèmes neurodégénératifs et développementaux. Dans le cadre de cette encéphalopathie, il a été suggéré que la réplication de l’ADN pouvait générer des fragments d’ADN cytosoliques, mais les mécanismes impliqués n'avaient pas été caractérisés.SAMHD1 est fréquemment muté dans le syndrome d’Aicardi-Goutières ainsi que dans certains cancers, mais son rôle dans l’étiologie de ces maladies était jusqu’à présent largement inconnu. Ce facteur de restriction du VIH possède une activité dNTPase impliquée dans la régulation des pools de nucléotides en G1, ainsi qu’une activité 3’-5’ exonucléase qui est, encore aujourd’hui, controversée.Le but de mon projet de thèse était de comprendre les mécanismes moléculaires responsables de l’inflammation dans les cellules déficientes pour SAMHD1.Nous montrons que de l’ADN cytosolique s’accumule dans les cellules déficientes pour SAMHD1, particulièrement en présence de stress réplicatif, activant la réponse interféron. Par ailleurs, SAMHD1 est important pour la réplication de l’ADN en conditions normales et pour le processing des fourches arrêtées, indépendamment de son activité dNTPase. De plus, SAMHD1 stimule l’activité exonucléase de MRE11 in vitro. Lorsque SAMHD1 est absent, la dégradation de l’ADN néosynthétisé est inhibée, ce qui empêche l’activation du checkpoint de réplication et entraine un défaut de redémarrage des fourches de réplication. De plus, la résection des fourches de réplication est réalisée par un mécanisme alternatif qui libère des fragments d’ADN dans le cytosol, activant la réponse interféron.Les résultats obtenus pendant ma thèse montrent, pour la première fois, un lien direct entre la réponse au stress réplicatif et la production d’interférons. Ces résultats ont des conséquences importantes dans notre compréhension du syndrome d’Aicardi Goutières et des cancers liés à SAMHD1. Par exemple, nous avons démontré que MRE11 et RECQ1 sont responsables de la production des fragments d’ADN qui déclenchent la réponse inflammatoire dans les cellules déficientes pour SAMHD1. Nous pouvons donc imaginer que bloquer l’activité de ces enzymes pourrait diminuer la production des fragments d’ADN et, in fine, l’activation de l’immunité innée dans ces cellules. Par ailleurs, la voie interférons joue un rôle essentiel dans l’efficacité thérapeutique de l’irradiation et de certains agents chimiothérapiques comme l’oxaliplatine. Moduler cette réponse pourrait donc avoir un intérêt beaucoup plus large en thérapie anti-tumorale. / DNA replication is an utterly complex process, implicating thousands of replication forks that progress along chromosomes. These forks frequently slow-down or stall when they encounter obstacles such as DNA secondary structures, proteins acting on chromatin or a lack of dNTPs. Such impediment on the progression of replication forks, termed replication stress, plays a major role in tumorigenesis. Intricate processes, still not entirely understood, are established to respond to this stress. For instance, nucleases such as DNA2 and MRE11 degrade nascent DNA at arrested forks to allow their restart.The interferon pathway is a defense mechanism against pathogens that detects non-self-nucleic acids in the cytosol to activate the innate immune response. However, DNA fragments originating from the metabolism of genomic DNA, such as DNA repair and retrotransposition, may also diffuse into the cytosol to activate this pathway. The Aicardi-Goutières Syndrome (AGS), a rare genetic disorder characterized by neurological and developmental defects is caused by chronic inflammation due to the over-production of type I IFNs. It has been proposed that DNA replication may generate cytosolic DNA fragments triggering this encephalopathy. However, the mechanisms involved have remained unexplored.SAMHD1 is frequently mutated in the Aicardi-Goutières Syndrome as well as in some cancers. However, its role in the etiology of these diseases remains poorly understood. This HIV-1 restriction factor has a dNTPase activity involved in the regulation of dNTP pools and a putative 3’-5’ exonuclease activity.The goal of my PhD thesis was to understand the molecular mechanisms responsible for inflammation induced by SAMHD1 deficiency.We show that cytosolic DNA accumulates in SAMHD1-deficient cells, especially in conditions of replication stress, activating the interferon response. In addition, we demonstrate that SAMHD1 is necessary for DNA replication and for the processing of arrested forks, independently of its dNTPase activity. SAMHD1 stimulates the exonuclease activity of MRE11 in vitro. In the absence of SAMHD1, nascent DNA degradation is inhibited, preventing replication checkpoint activation and fork restart. Besides, forks are aberrantly processed by an alternative pathway that generates cytosolic DNA fragments, thereby activating the interferon response.Altogether, we demonstrate for the first time a direct link between the DNA replication stress response and interferon production. This result has important consequences regarding our understanding of the Aicardi-Goutières Syndrome and cancers caused by SAMHD1 mutations, which could be translated into new therapeutic opportunities. For instance, we have shown that MRE11 and RECQ1 are responsible for the production of DNA fragments triggering the pro-inflammatory response in SAMHD1-deficient cells. Inhibiting these enzymes decreases the production of cytosolic nucleic acids and, consequently, reduces the activation of cell-autonomous innate immunity in SAMHD1-depleted cells. Accordingly, inhibiting these enzymes may as well decrease IFN production in AGS in vivo models caused by SAMHD1 mutations. The interferon pathway also plays a major role in tumorigenesis as well as in the clinical efficacy of irradiation and some chemotherapeutic agents such as oxaliplatin. Modulating this response may therefore have much broader implications in anti-cancer therapy.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第20559号 / 医博第4244号 / 新制||医||1022(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 高田 穣, 教授 萩原 正敏, 教授 松本 智裕 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Serial passaging of primary human fibroblasts leads to the formation of non-dividing senescent cells by a process termed replicative senescence. This tissue culture-based methodology is currently used as a model system to determine the underlying mechanisms of in vivo cellular aging and tumor suppression. Senescence is regarded as an alternative pathway to apoptosis, where cells undergo multiple changes in metabolic and cellular signaling pathways in order to prevent proliferation but still maintain a metabolically-active cell. Whether or not this model accurately reflects in vivo processes is presently controversial; however, replicative senescence is currently the most applicable model through which one can investigate the underlying causes of human cellular aging in the context of controlled environmental stress over time. This work was directed at understanding the molecular processes involved in replicative senescence with specific emphasis on the role of the mitochondria. A series of experiments were performed to assess changes during the induction of replicative senescence under conditions of low (3%) and high (20%) oxygen levels. Measurements were made at the transcriptional, protein, and metabolite levels. Microscopy wasalso utilized to monitor changes in mitochondrial morphology and volume. While previous studies have evaluated specific pathways and/or products; this work combines a more complete metabolomic, genomic, proteomic, and morphological picture of cells undergoing senescence and oxidative stress. Considering the low cell population densities of primary adherent fibroblasts and the subsequent low concentrations of small polar metabolites involved in glycolysis and the TCA cycle, methodologies needed to be developed in order to optimize metabolite extraction and liquid chromatography-mass spectrometric analysis. Protein kinase and transcriptional microarrays were also performed in order to quantify the changes in activated/deactivated signaling cascades as well as gene expression and relate these findings to metabolomic data. Mitochondrial dynamics of cells at different age time points and under different oxygen conditions were also assessed including mitochondrial size, shape, membrane potential, and percent volume per cell volume using confocal microscopy. The results obtained not only confirm the major pathways involved in senescence (p53/p21, PTEN/p27, and RTK/Raf/MAPK) but also provide evidence at both the transcriptional and protein levels for additional senescence-associated pathways. The majority of the changes observed were related to pathways involved in cellular stress, cell cycle control, and the survival response. Metabolic data suggested a –pooling effect– of glycolysis and TCA precursor molecules due to attenuation in enzyme function; this theory was also supported by an observed up regulation of gene expression as a compensatory mechanism. Mitochondria exhibited changes in membrane potential as well as volume and percent volume per cell which suggested compensatory hypertrophy and/or attenuation of mitochondrial fission processes. When the aforementioned analyses are tied together, a “theoretical model of senescence” can be formulated and is characterized by increased metabolic protein and associated metabolite levels due to attenuation in their respective enzyme function, resulting in increases in expression of their associated genes as a compensatory mechanism. / Ph. D.

Shortest-path network analysis was employed to identify novel genes that modulate longevity in the baker’s yeast Saccharomyces cerevisiae. Based upon a set of previously reported genes associated with increased life span, a shortest path network algorithm was applied to a pre-existing protein-protein interaction dataset in order to construct a shortest-path longevity network. To validate this network, the replicative aging potential of 88 single gene deletion strains corresponding to predicted components of the shortest path longevity network was determined. The 88 single-gene deletion strains identified by a network approach are significantly enriched for mutation conferring both increased and decreased replicative life span when compared to a randomly selected set of 564 single-gene deletion strains or to the current data set available for the entire haploid deletion collection. In addition, previously unknown longevity genes were identified, several of which function in a longevity pathway believed to mediate life span extension in response to dietary restriction. This study represents the first biologically validated application of a network construct to the study of aging and rigorously demonstrates, also for the first time, that shortest path network analysis is a potentially powerful tool for predicting genes that function as potential modulators of aging.

La réplication de l’ADN est un phénomène physiologique essentiel à la transmission du patrimoine génétique mais est aussi une source importante de stress endogène. Le stress réplicatif peut conduire à une instabilité génomique et a été mis en évidence à une étape très précoce du développement tumoral et de la sénescence. La recombinaison homologue (RH) est un processus de réparation qui permet la prise en prise en charge du stress réplicatif. De ce fait, un défaut de RH devrait permettre de révéler les stress réplicatifs endogènes. Ainsi, une progression ralentie des fourches de réplication a été observée dans des cellules déficientes pour la RH (RH-), et ce en absence de tout traitement exogène (Daboussi 2008). De plus, de nombreux travaux ont mis en évidence la présence de défauts mitotiques dans les cellules RH-, en absence de tout traitement exogène (Griffin 2000; Kraakman-van der Zwet 2002; Bertrand 2003; Daboussi 2005; Laulier 2011; Rodrigue 2013). L’origine de ces défauts mitotiques spontanés reste peu claire. En effet, la RH étant un processus préférentiellement actif au cours des phases S et G2, le lien avec la mitose reste à éclaircir. Cette thèse a pour but de comprendre l’impact du stress réplicatif très faible ou endogène sur les phases post-réplicatives du cycle cellulaire. Dans un premier temps, je me suis intéressée à l’impact de ce stress sur la mitose. Les résultats obtenus montrent que le traitement des cellules contrôle à de très faibles doses d’hydroxyurée (HU) n’affecte pas la progression dans le cycle cellulaire mais induit cependant une diminution de la vitesse de réplication, comparable à celle observée dans les cellules RH-. De plus le traitement des cellules contrôle à des faibles doses d’HU induit l’apparition de défauts mitotiques, notamment des centrosomes surnuméraires, à la même fréquence que dans les cellules RH- non traitées. Inversement, l’ajout de précurseurs de nucléotides dans les cellules RH- permet de supprimer la diminution de la vitesse de réplication ainsi que les centrosomes mitotiques surnuméraires. Ainsi, un stress réplicatif subtil, qui n’impacte pas de façon détectable la progression dans les phases S et G2 du cycle cellulaire, ni l’entrée en mitose, cause cependant des défauts mitotiques sévères. De façon importante, les centrosomes mitotiques surnuméraires peuvent entrainer des mitoses multipolaires, impactant ainsi l’ensemble du génome. Ces données mettent en évidence la connexion qui existe entre la réplication des chromosomes et leur ségrégation. Dans un second temps, j’ai étudié l’impact du stress réplicatif faible ou endogène en phase G2. Cette étude a été réalisée en utilisant des cellules RH-, ainsi qu’un modèle d’induction de faible stress réplicatif après traitement à très faible dose d’HU. La présence de foyers pRPA-Ser33 en phase G2 a été observée dans ces deux modèles, mettant en évidence des zones de stress réplicatif. Après traitement à très faible dose d’HU, nous observons également la présence en phase G2 de foyers 53BP1 et RAD51 qui colocalisent partiellement avec les foyers pRPA-Ser33. L’analyse en spectrométrie de masse après co-immunoprécipitation de la protéine 53BP1 en phase G2 a permis d’établir un lien avec des protéines impliquées dans le contrôle de l’assemblage du fuseau mitotique ainsi que dans le points de contrôle mitotique, étayant ainsi le lien entre le stress réplicatif et les défauts mitotiques. Pour finir, l’immunoprécipitation de la chromatine liée à la protéine pRPA-Ser33 en phase G2, suivie d’un séquençage (ChIPseq), a permis de révéler l’absence d’enrichissement au niveau des sites fragiles communs et de mettre en évidence un enrichissement au niveau des régions promotrices de certains gènes, notamment de gènes impliqués dans la régulation du cycle cellulaire et de la mort cellulaire. Ces résultats soulignent le lien entre le stress réplicatif très faible ou endogène et l’instabilité chromosomique, qui peut mener à l’initiation tumorale. / DNA replication is a physiological process, essential for genetic information transmission but DNA replication is also an important source of endogenous stress. Replicative stress can lead to genomic instability and has been reported in early-stage malignancies and senescence. Homologous recombination is a repair process which can handle replicative stress. Therefore, a defect in homologous recombination could reveal endogenous replicative stresses. Consistently, a slow down in replication fork progression has been observed in homologous recombination deficient (HR-) cells, in absence of any exogenous treatment (Daboussi et al. 2008). In addition, several studies have shown the presence of mitotic defects in HR- cells, in absence of any exogenous treatment (Griffin 2000; Kraakman-van der Zwet 2002; Bertrand 2003; Daboussi 2005; Laulier et al. 2011; Rodrigue 2013). The origin of these spontaneous mitotic defects is still unclear. Indeed, homologous recombination is preferentially active in S and G2 phases thus, the link with mitosis remains to be elucidated. The aim of this thesis is to understand the impact of a low or endogenous replicative stress on post-replicative phases. First, I studied the impact of a low or endogenous replicative stress on mitosis. Control cells were treated with very low hydroxyurea doses, that did not affected cell cycle progression but did slow down the replication fork progression to the same level than unchallenged HR- cells. Importanntly, exposure of the control cells to these low hydroxyurea doses generated the same mitotic defects, notably extra centrosomes, and to the same extent than in untreated HR- cells. Reciprocally, supplying nucleotide precursors to HR- cells suppressed both their replication deceleration and mitotic extra centrosome phenotypes. Therefore, subtle replication stress that does not impact S and G2 phase progression nor the entry in mitosis, nevertheless causes severe mitotic defects. Importantly, mitotic extra centrosome can lead to multipolar mitosis and then impact the whole genome stability. These data highlight the crosstalk between chromosome replication and segregation. Secondly, I studied the impact of low or endogenous replicative stress on G2 phase. This study was done using HR- cells as well as control cells treated with very low HU doses to induce a very low replicative stress. In both of these models, the presence of pRPA-Ser33 foci was observed in G2 phase, highlighting replicative stress regions. After very low HU treatement, we observed 53BP1 and RAD51 foci in G2 phase. These foci partially colocalized with pRPA-Ser33 foci in G2 phase. Mass spectrometry analyse after 53BP1 coimmunoprecipitation allowed to etablish a link between proteins involved in mitotic spindle assembly control and in mitotic checkpoint. These data support the link between replicative stress and mitotic defects. Lastly, the immmunoprecipitation of the chromatin interacting with pRPA-Ser33 in G2 phase, followed by sequencing (ChIPseq) allowed to reveal the absence of common fragile site enrichment and to highlight an enrichment at promoter regions of genes involved in cell cycle and cell death regulation. These data underline the link between very low or endogenous replicative stress and chromosomal instability, which can lead to tumorigenesis.

O FGF2 (Fibroblast Growth Factor 2) é um clássico fator peptídico de crescimento que ativa vias intracelulares de sinalização molecular promovendo a transição G0 &#8594; G1 e o comprometimento com o ciclo celular. Não surpreendentemente, seus papéis pró-tumoral e angiogênico estão bem caracterizados e estabelecidos na literatura. No entanto, um crescente corpo de evidências tem indicado que o FGF2 também pode exercer efeitos anti-tumorais in vitro e in vivo, em modelos murinos e também humanos. Neste contexto, nosso grupo publicou em 2008 que o FGF2 exerce um efeito antiproliferativo seletivo em células murinas malignas dependentes de alta atividade de K-Ras e H-Ras. Os genes ras compõem a família de oncogenes mais frequentemente mutada em tumores malignos humanos, alcançando aproximadamente 30% de todos os casos. O desenvolvimento de terapias contra tumores dependentes de Ras fracassou, apesar dos intensos esforços e investimentos desde a descoberta em 1982 de suas mutações ativadoras em múltiplos cânceres. O objetivo deste trabalho foi desvendar os mecanismos moleculares pelo quais o FGF2 inibe irreversivelmente a proliferação de células malignas dependentes da atividade de Ras, empregando como modelos experimentais a linhagem murina Y1 de células adrenocorticais, e 4 linhagens humanas derivadas de sarcomas de Ewing. Identificamos que o efeito citotóxico do FGF2 não se processa por um mecanismo novo e independente das viasproliferativas classicamente ativadas por fatores peptídicos de crescimento. Ao contrário, seu efeito tóxico é resultado de sinalização mitogênica exagerada decorrente de estimulação sustentada por FGF2. A ativação da via de MAPK, principal sinalização mitogênica intracelular, a níveis elevados e sustentados provoca estresse mitogênico, que se propaga para a fase S na forma de estresse replicativo. Nesta situação, a célula passa a depender exageradamente da sinalização protetora de ATR, de modo que a combinação de estimulação com FGF2 e inibição de ATR foi altamente letal para as células malignas dependentes de Ras empregadas neste trabalho. Também analisamos as bases moleculares de resistência a FGF2 exibida por células Y1 anteriormente selecionadas para resistir ao efeito tóxico do FGF2 (Y1FRs). Descobrimos que a pressão seletiva do FGF2 não teve efeito na expressão de seus receptores, mas provocou a eliminação de um dos dois cromossomos que portam a amplificação gênica de ras nesta linhagem, enquanto o segundo cromossomo foi mantido por ser a única fonte de genes ribossomais ativos. Suas cópias de ras, no entanto, mostraram-se transcricionalmente silenciadas. Além disso, as sublinhagens Y1FRs não expressam o principal RasGEF, GRP4, encontrado nas células parentais Y1, o que pode ter influenciado o surgimento do fenótipo resistente ao FGF2. As linhagens resistentes mostraram grande redução no número de cromossomos e aumento da frequência de fusões entre cromossomos não homólogos em relação às células parentais. / FGF2 (Fibroblast Growth Factor 2) is a classic peptide growth factor that activates intracellular molecular signaling pathways promoting the G0 &#8594; G1 transition and cell cycle commitment. Not surprisingly, its pro-tumor and angiogenic roles are well characterized and established in the literature. However, a growing body of evidence has indicated that FGF2 may also exert anti-tumor effects in vitro and in vivo in murine and human models. In this context, our group reported in 2008 that FGF2 exerts a selective antiproliferative effect in murine cells dependent on high activity of K-Ras and H-Ras. Ras genes make up the most frequently mutated oncogene family in human malignant tumors, reaching approximately 30% of all cases. The development of therapies against Ras-dependent tumors has failed despite intense efforts and investments since the discovery in 1982 of its activating mutations in multiple cancers. The objective of this work was to uncover the molecular mechanisms by which FGF2 irreversibly inhibits the proliferation of malignant cells dependent on Ras activity, using as experimental models the Y1 murine lineage of adrenocortical malignant cells and 4 human lineages derived from Ewing sarcomas. We showed that the cytotoxic effect of FGF2 did not involve novel cell cycle regulatory pathways; instead, this cytotoxic effect is a result of sustainedhyper mitogenic stimulation by FGF2. Activation of the KRas/MAPK pathway, the major intracellular mitogenic signaling, at high and sustained levels provokes mitogenic stress, which is propagated to S phase as replicative stress. In this situation, the cell dependence on the ATR protective signaling is enhanced, so that the combination of stimulation with FGF2 and inhibition of ATR was highly lethal for the Ras dependent malignant cells employed in this work. We also analyzed the molecular basis of FGF2 resistance exhibited by Y1 cells previously selected for resistance to FGF2. We found that the selective pressure of FGF2 had no effect on the expression of its receptors but promoted the elimination of one of the two marker chromosomes that carry the K-ras amplified copies, while the second chromosome was maintained because it is the only source of active ribosomal genes; however, its K-ras amplified copies were transcriptionally silenced. In addition, the Y1FRs sublines did not express the main RasGEF, GRP4, found in the parental Y1 cells, which might have played a role in the emergence of the FGF2-resistant phenotype. The resistant Y1FRs sublines showed a large reduction in chromosome numbers and increased frequency of fusions between non-homologous chromosomes in relation to parental cells.

O FGF2 (Fibroblast Growth Factor 2) é um clássico fator peptídico de crescimento que ativa vias intracelulares de sinalização molecular promovendo a transição G0 &#8594; G1 e o comprometimento com o ciclo celular. Não surpreendentemente, seus papéis pró-tumoral e angiogênico estão bem caracterizados e estabelecidos na literatura. No entanto, um crescente corpo de evidências tem indicado que o FGF2 também pode exercer efeitos anti-tumorais in vitro e in vivo, em modelos murinos e também humanos. Neste contexto, nosso grupo publicou em 2008 que o FGF2 exerce um efeito antiproliferativo seletivo em células murinas malignas dependentes de alta atividade de K-Ras e H-Ras. Os genes ras compõem a família de oncogenes mais frequentemente mutada em tumores malignos humanos, alcançando aproximadamente 30% de todos os casos. O desenvolvimento de terapias contra tumores dependentes de Ras fracassou, apesar dos intensos esforços e investimentos desde a descoberta em 1982 de suas mutações ativadoras em múltiplos cânceres. O objetivo deste trabalho foi desvendar os mecanismos moleculares pelo quais o FGF2 inibe irreversivelmente a proliferação de células malignas dependentes da atividade de Ras, empregando como modelos experimentais a linhagem murina Y1 de células adrenocorticais, e 4 linhagens humanas derivadas de sarcomas de Ewing. Identificamos que o efeito citotóxico do FGF2 não se processa por um mecanismo novo e independente das viasproliferativas classicamente ativadas por fatores peptídicos de crescimento. Ao contrário, seu efeito tóxico é resultado de sinalização mitogênica exagerada decorrente de estimulação sustentada por FGF2. A ativação da via de MAPK, principal sinalização mitogênica intracelular, a níveis elevados e sustentados provoca estresse mitogênico, que se propaga para a fase S na forma de estresse replicativo. Nesta situação, a célula passa a depender exageradamente da sinalização protetora de ATR, de modo que a combinação de estimulação com FGF2 e inibição de ATR foi altamente letal para as células malignas dependentes de Ras empregadas neste trabalho. Também analisamos as bases moleculares de resistência a FGF2 exibida por células Y1 anteriormente selecionadas para resistir ao efeito tóxico do FGF2 (Y1FRs). Descobrimos que a pressão seletiva do FGF2 não teve efeito na expressão de seus receptores, mas provocou a eliminação de um dos dois cromossomos que portam a amplificação gênica de ras nesta linhagem, enquanto o segundo cromossomo foi mantido por ser a única fonte de genes ribossomais ativos. Suas cópias de ras, no entanto, mostraram-se transcricionalmente silenciadas. Além disso, as sublinhagens Y1FRs não expressam o principal RasGEF, GRP4, encontrado nas células parentais Y1, o que pode ter influenciado o surgimento do fenótipo resistente ao FGF2. As linhagens resistentes mostraram grande redução no número de cromossomos e aumento da frequência de fusões entre cromossomos não homólogos em relação às células parentais. / FGF2 (Fibroblast Growth Factor 2) is a classic peptide growth factor that activates intracellular molecular signaling pathways promoting the G0 &#8594; G1 transition and cell cycle commitment. Not surprisingly, its pro-tumor and angiogenic roles are well characterized and established in the literature. However, a growing body of evidence has indicated that FGF2 may also exert anti-tumor effects in vitro and in vivo in murine and human models. In this context, our group reported in 2008 that FGF2 exerts a selective antiproliferative effect in murine cells dependent on high activity of K-Ras and H-Ras. Ras genes make up the most frequently mutated oncogene family in human malignant tumors, reaching approximately 30% of all cases. The development of therapies against Ras-dependent tumors has failed despite intense efforts and investments since the discovery in 1982 of its activating mutations in multiple cancers. The objective of this work was to uncover the molecular mechanisms by which FGF2 irreversibly inhibits the proliferation of malignant cells dependent on Ras activity, using as experimental models the Y1 murine lineage of adrenocortical malignant cells and 4 human lineages derived from Ewing sarcomas. We showed that the cytotoxic effect of FGF2 did not involve novel cell cycle regulatory pathways; instead, this cytotoxic effect is a result of sustainedhyper mitogenic stimulation by FGF2. Activation of the KRas/MAPK pathway, the major intracellular mitogenic signaling, at high and sustained levels provokes mitogenic stress, which is propagated to S phase as replicative stress. In this situation, the cell dependence on the ATR protective signaling is enhanced, so that the combination of stimulation with FGF2 and inhibition of ATR was highly lethal for the Ras dependent malignant cells employed in this work. We also analyzed the molecular basis of FGF2 resistance exhibited by Y1 cells previously selected for resistance to FGF2. We found that the selective pressure of FGF2 had no effect on the expression of its receptors but promoted the elimination of one of the two marker chromosomes that carry the K-ras amplified copies, while the second chromosome was maintained because it is the only source of active ribosomal genes; however, its K-ras amplified copies were transcriptionally silenced. In addition, the Y1FRs sublines did not express the main RasGEF, GRP4, found in the parental Y1 cells, which might have played a role in the emergence of the FGF2-resistant phenotype. The resistant Y1FRs sublines showed a large reduction in chromosome numbers and increased frequency of fusions between non-homologous chromosomes in relation to parental cells.