The genome of the hyperthermophile archaeon Pyrococcus furiosus encodes two transcription factor B (TFB) paralogs, one of which (TFB1) was previously characterized in transcription initiation. The second TFB (TFB2) is unusual in that it lacks recognizable homology to the archaeal TFB/eukaryotic TFIIB B-reader (also called the B-finger) motif. TFB2 functions, though poorly, in promoter-dependent transcription initiation. Domain swaps between TFB1 and TFB2 showed that the low activity of TFB2 is determined mainly by its N terminus. The low activity of TFB2 in promoter opening and transcription can be partially relieved by transcription factor E (TFE). The results indicate that the TFB N-terminal region, containing conserved Zn ribbon and B-finger motifs, is important in promoter opening and that TFE can compensate for defects in the N terminus through enhancement of promoter opening. Archaeal RNA polymerase requires two transcription factors for initiation: TBP, which binds to TATA boxes, and TFB, which binds TBP and DNA, recruits RNAP and helps initiate transcription. Archaeal TFBs usually contain a conserved B-reader sequence homologous to the eukaryotic B-reader motif in their N-terminal domains. This region is involved in the assembly of the transcription complex, promoter melting and in transcription start site determination but its position and orientation relative to promoter DNA during initiation is not clear. In this study the positioning of the TFB B-reader relative to DNA was determined by cross-linking using TFB variants substituted with photoactivatable unnatural amino acids. The results demonstrate that the B-reader is in close proximity to the transcription start site on the template but not the non-template strand in transcription initiation complexes. Furthermore, the position of the B-reader varies between closed and open promoter complexes, and between open promoter and early initiation complexes. Thus the archaeal B-reader sequence is poised to interact with promoter DNA in a dynamic fashion, and is likely playing a role in positioning the template-strand in an open pre-initiation complex.

The hyperthermophile archaeon, Pyrococcus furiosus encodes two eukaryotic TFIIB family proteins, TFB1 and TFB2. TFB1 is very similar to TFIIB in terms of sequence homology and function, whereas TFB2 is unusual as it is missing highly conserved sequences in its N-terminal domain that are present in TFIIB and TFB1. Despite this, TFB2 is effective in transcription process, albeit with lower efficiency compared to TFB1. Other archaea also contain multiple TFBs, but unlike Pyrococcus furiosus TFB2, these multiple TFBs have higher sequence homology to each other and have similar transcription efficiencies. Photochemical cross-linking experiments have shown that the B-reader of TFB in archaea and TFIIB in eukaryotes is close to transcription start site and is very important in RNAP recruitment to promoter DNA and transcription start site selection. Thus the lack of the highly conserved B reader region in P. furiosus TFB2 presents the opportunity to further study the functional importance of this region. In this study several amino acids in N-terminal domain of TFB2 were mutated with photoactivable unnatural amino acid p-benzoyl L- phenylalanine (pBpa) and the proximity of TFB2 relative to DNA was determined by photochemical cross-linking experiments. The results showed that TFB2 interacts with DNA near the TATA box via its C-terminal domain, and interacts with both strands of DNA near the transcription start site via its divergent B-reader and the B-linker sequences. The B-reader loop region is close to transcription start site and interacts with the transcribed strand of promoter DNA while the B-linker strand cross-links with the non-transcribed strand. Some of the amino acids in between the B-reader loop and the B-linker strand region in TFB2 are seen to cross-link both the transcribed and the non-transcribed strand. Thus, despite the absence of strong homology to conserved B-reader and B-linker sequences, TFB2 is likely to interact with DNA in the transcription bubble and facilitate in transcription initiation.

With the critical role of aberrantly active Signal Transducer and Activator of Transcription (Stat) 3 protein in many human cancers, selective small-molecule inhibitors targeting the dimerization event which is required for stat3 activation, would be valuable as therapeutic agents. And the inhibitors will be useful chemical probes to clarify the complex biological functions of Stat3. By computational and structural analyses of the interaction between Stat3 and the lead dimerization disruptor, S3I-201, we have designed a diverse set of analogs. One of the most active analogs, S3I-201.1066 is derived to contain a cyclo-hexyl benzyl moiety on the amide nitrogen, which increases the binding to the Stat3 SH2 domain. Evidence is presented from in vitro biochemical and biophysical studies that S3I-201.1066 directly interacts with Stat3 or the SH2 domain, with an affinity (K[subscript D]) of 2.74 [micrometer], and disrupts the binding of Stat3 to the cognate pTyr-peptide, GpYLPQTV-NH2, with an IC₅₀ of 23 [micrometer]. Moreover, S3I-201.1066 selectively blocks the association of Stat3 with the epidermal growth factor receptor (EGFR), and inhibits Stat3 tyrosine phosphorylation and nuclear translocation in EGF-stimulated mouse fibroblasts. In cancer cells that harbor aberrant Stat3 activity, S3I-201.1066 inhibits constitutive Stat3 DNA-binding and transcriptional activities. By contrast, S3I-201.1066 has no effect on Src activation or the EGFR-mediated activation of the Erk1/2MAPK pathway. S3I-201.1066 selectively suppresses the viability, survival, and malignant transformation of the human breast and pancreatic cancer lines and the v-Src-transformed mouse fibroblasts harboring persistently active Stat3. Treatment with S3I-201.1066 on malignant cells harboring aberrantly active Stat3 down regulated the expression of c-Myc, Bcl-xL, Survivin, matrix metalloproteinase 9, and VEGF, which are known Stat3-regulated genes important in diverse tumor processes. The in vivo administration of S3I-201.1066 induced significant anti-tumor response in mouse models of human breast cancer, which correlates with the inhibition of constitutively active Stat3 and the suppression of known Stat3-regulated genes. Further computer-aided lead optimization derives higher-affinity (K[subscript D], 504 nM), orally bioavailable Stat3 SH2 domain-binding ligand, BP-1-102 as a structural analog of S3I-201.1066. The most significant modification is the pentafluorobenzene sulfonamide component of BP-1-102, which permits accessibility of a third sub-pocket of the Stat3 SH2 domain surface. BP-1-102-mediated inhibition of aberrantly-active Stat3 in human pancreatic cancer, Panc-1, breast cancer, MDA-MB-231, and prostate (DU145) cancer cells and in the mouse transformed fibroblasts harboring aberrantly-active Stat3. It also disrupts Stat3-NF[kappa]B cross-talk and suppresses the release of granulocyte colony-stimulating factor, soluble intercellular adhesion molecule-1, macrophage-migration-inhibitory factor/glycosylation-inhibiting factor, interleukin-1 receptor antagonist and the serine protease inhibitor (serpin) protein 1, and the expression of c-Myc, Cyclin D1, Bcl-xL, Survivin, and vascular endothelial growth factor expression in vitro and in vivo. Inhibition of tumor cell-associated constitutively-active Stat3 further suppresses focal adhesion kinase and paxillin induction, enhances E-cadherin expression, and down-regulates Kruüppel-like factor 8 expression. Consequently, BP-1-102 selectively suppresses anchorage-dependent and independent growth, survival, migration and invasion of Stat3-dependent tumor cells in vitro. Intravenous or oral gavage delivery of BP-1-102 furnishes micromolar or microgram levels in tumor tissues and inhibits growth of mouse xenografts of human breast and lung tumors. Computer-aided lead optimization has therefore derived a more suitable small-molecule inhibitor as a drug candidate. Our studies of the Stat3 SH2 protein surface and of the interactions between lead agents and the SH2 domain provided significant data to facilitate the structural optimization. From S2I-201 to S3I-201.1066 and to BP-1-102, we note the substantial gain in potency and efficacy, and the pharmacokinetic improvements. The oral bioavailability of BP-1-102 represents a substantial advancement in the discovery of small-molecule Stat3 inhibitors as novel anticancer agents.

La chromatine eucaryote, contenant l’ADN et de nombreuses protéines de liaison, subit une compaction dynamique et fonctionnelle à de multiples échelles, nécessaire pour la régulation de nombreux processus biologiques comme l’expression génique. Afin de définir et maintenir les fonctions cellulaires, les protéines de la régulation transcriptionnelle et de la régulation de la structure chromatinienne agissent de concert pour orchestrer les programmes d’expression génique des cellules. Les facteurs de transcription opèrent de manière combinée et hiérarchique au niveau de nombreux éléments régulateurs, dont le fonctionnement est complexe et intégré, capables de générer de larges boucles topologiques pour réguler spécifiquement un promoteur cible à un moment précis. Le co-activateur transcriptionnel Mediator sert de centre d’interprétation, en connectant physiquement les régulateurs de la transcription à la machinerie transcriptionnelle, pour générer une réponse calibrée. Le complexe de maintenance de la structure des chromosomes, Cohesin, est impliqué dans la formation et la stabilisation des connexions génomiques à l’échelle de nombreuses structures chromatiniennes tri-dimensionnelles dont la caractérisation fonctionnelle commence à être explorée. Ensemble, les facteurs de transcription, Mediator et Cohesin contrôlent l’expression des programmes responsables du maintien de l’identité cellulaire. Les cellules cancéreuses présentent de nombreuses dérégulations au niveau transcriptionnel, et donc un programme d’expression aberrant. Nous avons démontré que les mécanismes de régulation qui contrôlent les cellules cancéreuses sont conservés, et proposons une stratégie qui permette de révéler les facteurs clefs dans la progression tumorale. Nous avons appliqué cette stratégie à la problématique de la résistance endocrinienne dans la progression du cancer du sein hormono-dépendant. Les résultats obtenus suggèrent que le complexe transcriptionnel AP-1 pourrait être impliqué dans l’acquisition et/ou le maintien de la résistance, en réponse aux pressions de sélection induites par les traitements hormonaux. Nous proposons une adaptation progressive et agressive des cellules cancéreuses par re-hiérarchisation des facteurs clefs qui contrôlent sa croissance. / Human chromatin, that contain both DNA and numerous binding proteins, is the target of a dynamic and functional multi-scaled compaction, which leads to the regulation of biologic processes as gene expression. In order to define and maintain cellular functions, transcriptional and structural regulatory proteins act together to orchestrate the different genes expression programs. Transcription factors function in a combinatorial and hierarchical manner on regulatory elements within the genome, which are able to generate large topological loops to specifically regulate a target promoter in the right temporal frame. The general co-activator Mediator functions as a center for proper transduction of the transcriptional input, physically connecting regulatory proteins to the transcriptional machinery, to generate a calibrated biological response. Cohesin is implicated into the formation and stabilization of genomic connections at multi-scaled tri-dimensional resolution, which functional features are beginning to be elucidated. Together, transcription factors, Mediator and Cohesin control expression programs that enable the maintenance of cellular identities. Cancerous cells often show deregulations at the transcriptional level, which lead to aberrant expression programs. We demonstrated that regulatory mechanisms, controlling transcription in cancerous cells, obey to the same rules that in normal cells and are conserved, then enable the characterization of key transcription factors that drive cancer progression. We applied this discovery to hormonal resistance in breast cancers. Our results suggest that AP-1 family could be involved into the acquisition of this more aggressive phenotype, by transcriptionally bypassing the drug effects. We proposed that a model for aggressiveness in cancer cells could be through their adaptation to transcriptional treatments, leading to a modulation of key important transcription factors driving transcriptional programs within the cells.

published_or_final_version / abstract / Biochemistry / Master / Master of Philosophy

Ovarian carcinoma is the most lethal gynecological malignancy. The high relapse and mortality rates are attributable to late diagnosis and development of drug resistance. Identifying novel prognostic and therapeutic targets for ovarian carcinoma is crucial for improving patients' long-term survival rate. Forkhead box protein M1 (FOXM1), which is a widely studied member of the FOX superfamily of proteins, participates in cell proliferation and apoptosis affecting the developmental function of many organs. Recently, there has been emerging evidence supporting the biological significance of FOXM1 in carcinogenesis. Overexpression of FOXM1 has been reported in multiple human malignancies including primary breast cancer, lung cancer, prostate cancer, etc. However, whether FOXM1 participates in the development of ovarian cancer, with emphasis on the association with clinicopathological parameters and chemoresistance, remains unknown. This study aims at elucidating the functional role of FOXM1 in the tumorigenesis of ovarian cancer. Immunohistochemical study showed higher nuclear FOXM1 expression was significantly associated with advanced stages of ovarian cancer (P=0.035). Though not reaching statistical significance, FOXM1 overexpression displayed association with serous histologic subtype, high grade cancers (poor differentiation) and chemoresistance. Patients with a low FOXM1 level had a significantly longer overall (P=0.019) and disease-free survival (P=0.014) than those with high FOXM1 expression. Multivariate progression analysis established high expression of FOXM1, advanced cancer stages and poor histological differentiation (high grade) as independent prognostic factors for short overall and disease-free survival. Consistently, in vitro Transwell assays demonstrated transient knockdown of FOXM1 was capable of reducing SKOV-3 migration and invasion. Furthermore, paclitaxel treatment down-regulated FOXM1 expression in the sensitive cell line but not the resistant one. Immunofluorescence and flow cytometric analyses demonstrated FOXM1 knockdown could enhance paclitaxel-mediated mitotic catastrophe in ovarian cancer cells. Recent attention has been drawn to the oncogenic roles of kinesin-like protein KIF2C and p21-activated kinase 4 (PAK4) in human cancers. Interestingly, the expressions of KIF2C and PAK4 altered in a similar pattern to FOXM1 expression upon paclitaxel treatment by displaying down-regulation only in the paclitaxel sensitive cell line but not the resistant one. FOXM1 silencing, qPCR, luciferase reporter assay and chromatin immunoprecipitation confirmed KIF2C and PAK4 to be novel transcriptional targets of FOXM1. Clonogenic assay showed KIF2C knockdown could re-sensitize resistant cell line to paclitaxel treatment. Flow cytometry demonstrated KIF2C silencing was able to increase the number of cells blocked at G2/M cell cycle phase in sensitive cell line and raise the number of apoptotic cells in resistant cell line. Up-regulations of miR-590 and miR-370 were also observed in a panel of drug resistant ovarian and breast cancer cell lines. While ectopic expression of miR-590 reduced FOXM1 expression, FOXM1 also seemed to be able to regulate the expression of miR-590. In summary, this study showed overexpression of FOXM1 in ovarian cancer correlated with poor survival of patients and paclitaxel resistance. KIF2C and PAK4 were identified as novel transcriptional targets of FOXM1 implicated in chemoresistance. / published_or_final_version / Pathology / Doctoral / Doctor of Philosophy

L’expression d’un gène donné est généralement le résultat de la dualité qui existe entre l’activation et la répression transcriptionnelle de ce gène. Au laboratoire, nous tentons de mieux comprendre la régulation de la transcription génique, et c’est dans ce cadre que nous étudions certaines protéines qui appartiennent à la famille de facteurs de transcription Ets. Ces derniers sont caractérisés par un domaine de liaison à l’ADN hautement conservé, le domaine ETS, qui se lie sur les promoteurs de leurs gènes cibles au niveau de sites comportant le motif central 5’- GGAA/T -3’. Certaines de ces protéines ont été montrées comme étant impliquées dans des processus du développement normal et cancéreux. Dans la première partie de ce travail, nous avons étudié le facteur transcriptionnel Fev dont l’expression est restreinte au noyau du raphé dans le cerveau, à la prostate et à l’intestin grêle. Nous avons participé à la caractérisation fonctionnelle de ce facteur permettant de le définir comme répresseur transcriptionnel. Plus particulièrement, nous avons identifié la partie carboxy-terminale riche en résidus alanine comme étant impliquée dans la répression. Afin de mieux comprendre le mécanisme moléculaire qui régit la répression induite par Fev, nous avons tenté d’identifier les partenaires protéiques impliqués dans ce processus transcriptionnel. Dans un premier temps, nous avons montré que Fev interagit physiquement avec les co-répresseurs transcriptionnels à activité histone désacétylase HDAC1 et HDAC3. Aussi, nous proposons de définir le rôle biologique de cette interaction. Par la suite, nous avons utilisé le système de criblage de banques par « double hybride en levures ». En utilisant comme appât soit Fev, soit sa partie carboxy-terminale, nous avons isolé plusieurs candidats interacteurs, dont la protéine DP103 qui est impliquée dans la régulation transcriptionnelle induite par d’autres facteurs de transcription de la famille Ets. Après avoir montré par co-immunoprécipitation que Fev interagit avec DP103, nous tentons de mettre en évidence la fonctionnalité de cette interaction. Dans la seconde partie de ce travail, nous avons étudié Erm, un activateur transcriptionnel de la famille Ets, qui est exprimé dans certains types de tumeurs, telles que les cancers mammaires métastatiques, et qui y régule l’expression de métalloprotéases. Ce facteur joue aussi un rôle régulateur dans les lymphocytes CD4+ T helper de type 1 (Th1) via l’interleukine-12. Néanmoins, les cibles ainsi que le rôle de Erm ne sont pas encore clairement identifiés dans les lymphocytes. Dans cette partie du travail, nous avons initié une étude sur les voies de signalisation impliquées dans la régulation transcriptionnelle de Erm. Nous avons montré que dans la lignée cellulaire Molt4 d’origine lymphoblastique ce facteur de transcription est la cible de la cascade de signalisation impliquant la famille des protéines kinases C (PKC). Grâce à l’utilisation d’inhibiteurs spécifiques des différentes sous-familles des PKC, nous avons montré que la transcription de Erm est régulée par les PKC conventionnelles. Aussi, après avoir isolé un fragment de 0,5 Kpb du promoteur de Erm, en amont de l’exon 1a, nous avons identifié une région régulatrice qui est activée par la voie des PKC. Ainsi, les approches que nous avons développées dans ce travail nous ont permis de progresser dans la caractérisation des facteurs de transcription Fev et Erm.

No description available.

No description available.

No description available.