The centrosome is the main microtubule-organising centre in animal cells; important to assemble a bipolar mitotic spindle ensuring proper chromosome segregation and genomic stability. Whereas correct centrosome number (1-2) is tightly maintained in normal cells, cancer cells usually have an increased number of centrosomes (>2), termed centrosome ampli cation. Centrosome ampli cation has been correlated with aneuploidy, increased tumour grade, chemoresistance and overall poor prognosis. Cancer cells primarily adapt to supernumerary centrosomes by clustering them into two poles resulting in a `normal' pseudo-bipolar mitosis. Undermining centrosome clustering is a potential target for cancer-speci c treatment. Indeed, depleting the kinesin HSET has already been shown to speci cally kill cancer cells by impairing the centrosome clustering mechanism. However, it is unclear whether this process requires adaptation or it is inherent to all cell types. Using a panel of non-transformed cell lines, we observed that cells expressing Ecadherin have ine cient clustering mechanisms compared to cell lines without E-cadherin. Loss of E-cadherin (siRNA/CRISPR) promotes centrosome clustering and survival of epithelial cells with multiple centrosomes. In addition, loss of DDR1, involved in regulating cortical contractility downstream of E-cadherin, increases centrosome clustering in epithelial cells. Using Atomic Force Microscopy we con rmed that indeed loss of E-cadherin leads to increased cortical contractility in mitotic cells. Inhibition of actomyosin contractility prevents e cient clustering in cells that do not express E-cadherin, further suggesting that it is important for this process. Loss of E-cadherin and DDR1 is strongly correlated with high levels of centrosome ampli cation in breast cancer cell lines suggesting that these changes are an important adaptation mechanism to centrosome amplification.

Bien que divers mécanismes coopèrent pour empêcher les erreurs lors de la synthèse des protéines chez les bactéries, des erreurs traductionnelles de type « frameshift » (ETFs) ou « faux-sens » peuvent avoir lieu. En particulier, les ETFs ont été détectées à de faibles niveaux lors de la phase de croissance exponentielle et à des niveaux plus élevés durant la phase de croissance stationnaire chez Escherichia coli et Bacillus subtilis. Ces observations ont conduit les chercheurs à revoir le rôle de la "réponse stringente" dans la survenue des ETFs, qui constitue l’un des mécanismes clé de l'adaptation bactérienne aux changements nutritionnels. Elle découle de l'interaction entre un ribosome en cours de traduction et la protéines RelA/SpoT ce qui permet de détecter les ARNs de transfert (ARNts) non chargés et résulte en la production d'une molécule appelée (p)ppGpp . Dans une souche mutante relA incapable de synthétiser le (p)ppGpp, les ETFs sont fortement augmentées.Dans ce contexte, notre objectif principal a été de revisiter le rôle de la réponse stringente dans le contrôle des erreurs traductionnelles et de clarifier le rôle des deux métabolites antagonistes GTP et (p)ppGpp. Par exemple, le GTP stimule l'initiation de la traduction (en ciblant le facteur d'initiation IF2) alors que le (p)ppGpp inhibe l'initiation de la traduction (en rentrant en concurrence avec le GTP pour se fixer sur IF2).A cette fin, nous avons utilisé le modèle des bactéries à Gram positif B. subtilis, conçu trois systèmes rapporteurs distincts pour détecter les ETFs et construit une souche incapable de synthétiser du (p)ppGpp (appelée "(p)ppGpp0"). Nous avons observé qu'au cours de la croissance dans des milieux pauvres, les ETFs augmentent en l'absence de (p)ppGpp durant la phase exponentielle et que, contrairement à la souche sauvage, la souche (p)ppGpp0 présente un pic d’ETFs en milieu riche pendant la transition à la phase stationnaire. En contrôlant les niveaux intracellulaires de GTP dans la souche (p)ppGpp0, nous avons montré que l'abondance de GTP est le facteur qui déclenche l'apparition des ETFs. Néanmoins, après une "faible" induction de la biosynthèse du GTP conduisant à des taux de croissance sous-optimaux, le niveau d’ETFs forme toujours un pic lors de la transition vers la phase stationnaire, ce qui montre que le mode d'action du (p)ppGpp pour prévenir l'apparition des ETFs ne repose pas uniquement sur son action inhibitrice de la biosynthèse du GTP. Nous nous sommes alors concentrés sur l'effet inhibiteur du (p)ppGpp sur IF2 et avons mimé son action en injectant des drogues connues pour inhiber l'initiation de la traduction. Nous avons ainsi démontré qu'en réduisant l'initiation de la traduction lors de l'épuisement des aminoacyl-ARNts, la souche "(p)ppGpp0" est capable de contrôler de façon optimale le taux d’ETFs lors de la transition vers la phase stationnaire.Dans une deuxième partie, nous avons étudié comment la transcription et la traduction sont affectées par les variations du niveau de GTP et de (p)ppGpp. Nous avons observé que les gènes possédant un "+1" de transcription (TSS, « transcription start site ») composé de deux guanines (gènes artificiels et ARNs ribosomaux) ont vu leur taux de transcription positivement corrélés au taux de croissance à l'inverse des gènes possédant un TSS composé de deux adénines. Cette différence est encore plus prononcée pour la souche (p)ppGpp0 cultivée en milieu riche lors de l'ajout de guanosine (ce qui conduit à un niveau élevé de GTP).En conclusion, nous avons démontré que le (p)ppGpp contrôle le niveau d'erreurs traductionnelles lors de la croissance en régime permanent en abaissant les niveaux de GTP et lors d’un changement nutritionnel en inhibant spécifiquement l'initiation de la traduction, assurant une allocation parcimonieuse des ressources au sein de la bactérie. / Even though diverse mechanisms cooperate to prevent protein synthesis errors in bacteria, missense and translational frameshift errors (TFEs) can occur . In particular, TFEs were detected at low levels in the exponential growth phase and at higher levels in the stationary phase in both Escherichia coli and Bacillus subtilis. This observation led researchers to revisit the role of the “stringent response” in the occurrence of TFEs since it is the key mechanism involved in the bacterial adaptation to nutritional downshifts. It relies on the interaction between the RelA/SpoT proteins and the translating ribosomes, which leads to the detection of uncharged tRNAs and to the production of an alarmone called (p)ppGpp. In a relA mutant strains unable to synthesize (p)ppGpp, translational errors are highly increased.In this context, the main goal of our work was to revisit the role of the stringent response in the translational error control and to clarify the role of the two key, antagonistic metabolites GTP and (p)ppGpp. Indeed, while GTP enhances translation initiation (targeting the initiation factor IF2) and elongation (targeting the elongation factor EF-Tu) , (p)ppGpp inhibits GTP biosynthesis (reducing the enzyme activity of Gmk, HprT and GuaB) and translation initiation (competing with GTP on IF2).For this purpose, we used the Gram positive model bacterium B. subtilis, designed three distinct reporter systems to detect TFEs and built a strain unable to synthesize (p)ppGpp (called “(p)ppGpp0”). We observed that during growth in poor media TFEs were increased in the absence of (p)ppGpp in the exponential phase (i.e. steady-state growth) and that by contrast to the wild type, the (p)ppGpp0 strain exhibited a TFE burst during the transition in rich medium to the stationary phase. By controlling intracellular levels of GTP in the (p)ppGpp0 strain, we showed that GTP abundance is the trigger factor of TFEs occurrence. Nevertheless, upon a "weak" induction of GTP biosynthesis leading to sub-optimal growth rates, the TFEs rate still peaked during the transition to the stationary phase, which demonstrated that the mode of action of (p)ppGpp to prevent TFEs occurrence did not only rely on its inhibition of GTP biosynthesis. We then focused on the (p)ppGpp inhibitory effect on IF2 and mimicked its action by injecting drugs known to inhibit translation initiation. Hence, we demonstrated that by reducing translation initiation (injecting drugs) upon aminoacyl-tRNAs depletion (p)ppgGp0 wild-strain type cells is are able to optimally control the rate of TFEs in the transition to the stationary phase. The same conclusion is obtained even in presence of a high GTP level.In a second part, we studied how transcription and translation are affected by variations in GTP and (p)ppGpp abundances. We observed that genes possessing a transcription start site (TSS) made of two guanines were more importantly transcribed at higher growth rates than genes possessing a TSS made of two adenines. This difference was even more pronounced for (p)ppGpp0 strains grown in rich medium upon guanosine addition (leading to a high level of GTP). Moreover, the ribosomal RNAs (rrns; for which the TSS is a guanine) synthesis level seemed to be positively correlated to GTP levels during exponential growth in poor and rich media as observed by the modulation of GTP biosynthesis.In conclusion, we demonstrated that (p)ppGpp controls the occurrence of translational errors during steady-state growth by decreasing GTP levels and during a nutritional downshift by specifically inhibiting translation initiation ensuring a parsimonious , which also globally affects resource allocation.

No description available.

Linhagens celulares humanas têm atraído grande interesse devido a sua capacidade de glicosilar proteínas de maneira mais semelhante às proteínas nativas humanas, reduzindo o potencial de respostas imunológicas contra epítopos não humanos. No entanto, por se tratar de uma aplicação recente, essas células ainda não foram extensamente caracterizadas e cultivadas em condições reprodutíveis da escala industrial, ou seja, em suspensão e em meios de cultura livres de soro fetal bovino (SFB). Em função disso, o objetivo principal deste trabalho foi estabelecer culturas livres de SFB e em suspensão para as linhagens celulares humanas SK-Hep-1, HepG2 e HKB-11, que têm despertado grande interesse devido ao potencial de produção de proteínas recombinantes. Para isso, quatro formulações comerciais livres de SFB foram avaliadas. As células que apresentaram bons resultados na adaptação aos meios realizada em garrafas estáticas foram então adaptadas para crescimento em suspensão. Foi possível realizar a adaptação satisfatória da célula HKB-11 ao meio FreeStyle e da célula SK-Hep-1 ao meio SFMII bem como a criopreservação das mesmas também em condições livres de SFB. A caracterização cinética das células adaptadas mostrou que a célula HKB-11 apresentou concentração celular quatro vezes superior a da célula SK-Hep-1 (8,6x106 e 1,9x106 células/mL, respectivamente) e apresentou crescimento celular durante 18 dias em cultura. A velocidade específica de crescimento máxima (?max) foi semelhante nas duas células (0,0159 h-1 para a HKB-11 e 0,0186 h-1 para SK-Hep-1). A limitação do crescimento das células adaptadas não parece estar associada à exaustão de glicose e glutamina, tampouco à formação de lactato em concentrações inibitórias. Todavia, para ambos os casos, foi observada produção de amônia em concentrações consideradas inibitórias (2 - 5 mM). De maneira geral, foi possível estabelecer culturas celulares em condições compatíveis com o desenvolvimento de um bioprocesso reprodutível, seguro e em concordância com as boas práticas de fabricação. / Human cell lines have attracted great interest since they are capable of producing glycosylated proteins in a more similar way to native human proteins, reducing the potential for immune responses against non-human epitopes. However, these human cell lines have not been extensively characterized and cultured in large scale and in serum-free suspension conditions. As a result, the main objective of this work was to adapt three human cell lines: SK-Hep-1, HepG2 and HKB-11 to serum-free suspension cultures, since they are promising systems of recombinant protein expression. For this task, four commercial serum-free media were tested. Adapted cell lines in T-flasks were further adapted to suspension cultures. Results showed that both HKB-11 and SK-Hep-1 were adapted to serum-free suspension cultures in FreeStyle and SFMII, respectively and were cryopreservated in serum-free formulations. Kinetic characterization showed that HKB-11 cell concentration was four times higher than SK-Hep-1 cell (8,6x106 and 1,9x106 cells/ml, respectively) and showed cell growth in culture over 18 days. The maximum specific growth rate (?max) was similar for both cell lines (0,0159 h-1 to HKB-11 and 0,0186h-1 to SK-Hep-1). Growth limitation of adapted human cell lines does not seem to be associated with depletion of glucose and glutamine, nor with the formation of lactate in inhibitory concentrations. However, in both cases, ammonia production achieved inhibitory concentrations (2 - 5 mM). In general, it was possible to establish human cell cultures that are compatible with reproducible and safe bioprocess conditions and in compliance with good manufacturing practices.

Linhagens celulares humanas têm atraído grande interesse devido a sua capacidade de glicosilar proteínas de maneira mais semelhante às proteínas nativas humanas, reduzindo o potencial de respostas imunológicas contra epítopos não humanos. No entanto, por se tratar de uma aplicação recente, essas células ainda não foram extensamente caracterizadas e cultivadas em condições reprodutíveis da escala industrial, ou seja, em suspensão e em meios de cultura livres de soro fetal bovino (SFB). Em função disso, o objetivo principal deste trabalho foi estabelecer culturas livres de SFB e em suspensão para as linhagens celulares humanas SK-Hep-1, HepG2 e HKB-11, que têm despertado grande interesse devido ao potencial de produção de proteínas recombinantes. Para isso, quatro formulações comerciais livres de SFB foram avaliadas. As células que apresentaram bons resultados na adaptação aos meios realizada em garrafas estáticas foram então adaptadas para crescimento em suspensão. Foi possível realizar a adaptação satisfatória da célula HKB-11 ao meio FreeStyle e da célula SK-Hep-1 ao meio SFMII bem como a criopreservação das mesmas também em condições livres de SFB. A caracterização cinética das células adaptadas mostrou que a célula HKB-11 apresentou concentração celular quatro vezes superior a da célula SK-Hep-1 (8,6x106 e 1,9x106 células/mL, respectivamente) e apresentou crescimento celular durante 18 dias em cultura. A velocidade específica de crescimento máxima (?max) foi semelhante nas duas células (0,0159 h-1 para a HKB-11 e 0,0186 h-1 para SK-Hep-1). A limitação do crescimento das células adaptadas não parece estar associada à exaustão de glicose e glutamina, tampouco à formação de lactato em concentrações inibitórias. Todavia, para ambos os casos, foi observada produção de amônia em concentrações consideradas inibitórias (2 - 5 mM). De maneira geral, foi possível estabelecer culturas celulares em condições compatíveis com o desenvolvimento de um bioprocesso reprodutível, seguro e em concordância com as boas práticas de fabricação. / Human cell lines have attracted great interest since they are capable of producing glycosylated proteins in a more similar way to native human proteins, reducing the potential for immune responses against non-human epitopes. However, these human cell lines have not been extensively characterized and cultured in large scale and in serum-free suspension conditions. As a result, the main objective of this work was to adapt three human cell lines: SK-Hep-1, HepG2 and HKB-11 to serum-free suspension cultures, since they are promising systems of recombinant protein expression. For this task, four commercial serum-free media were tested. Adapted cell lines in T-flasks were further adapted to suspension cultures. Results showed that both HKB-11 and SK-Hep-1 were adapted to serum-free suspension cultures in FreeStyle and SFMII, respectively and were cryopreservated in serum-free formulations. Kinetic characterization showed that HKB-11 cell concentration was four times higher than SK-Hep-1 cell (8,6x106 and 1,9x106 cells/ml, respectively) and showed cell growth in culture over 18 days. The maximum specific growth rate (?max) was similar for both cell lines (0,0159 h-1 to HKB-11 and 0,0186h-1 to SK-Hep-1). Growth limitation of adapted human cell lines does not seem to be associated with depletion of glucose and glutamine, nor with the formation of lactate in inhibitory concentrations. However, in both cases, ammonia production achieved inhibitory concentrations (2 - 5 mM). In general, it was possible to establish human cell cultures that are compatible with reproducible and safe bioprocess conditions and in compliance with good manufacturing practices.