Une analyse multiple des mutations, de l’expression des transcrits et de l’épissage alternatif a été réalisée dans des biopsies de lésions colorectales, correspondant à des stades variés de transformation, afin de rechercher des altérations qui caractériseraient la transformation de la muqueuse normale en adénome, puis en adénocarcinome. Cette analyse est basée sur l’utilisation de différentes technologies de puces à ADN. Des altérations spécifiques à chaque type de lésions colorectales ont été mises en évidence, démontrant que les adénomes et les adénocarcinomes sont des entités distinctes. Cependant, des altérations communes ont aussi été identifiées, confirmant que l’adénome est un état transitoire avant l’adénocarcinome. Une sélection clonale et des effets environnementaux sont sans doute à l’origine de la progression des adénomes en adénocarcinomes. Des voies de signalisation cellulaire caractéristiques de cette transformation ainsi qu’une classification des lésions colorectales ont été recherchées. Une signature de 40 transcrits a été identifiée, qui pourrait permettre de prédire la transformation des adénomes en adénocarcinomes. Des événements d’épissage alternatif ont aussi été détectés dans les adénomes, suggérant l’implication, à un stade précoce, de ces altérations dans le processus de cancérisation. Enfin, une puce à ADN « à façon » a été élaborée, qui s’inscrit dans une perspective d’appui à la mise en oeuvre de thérapeutiques anticancéreuses. Elle permet en effet d’analyser l’épissage alternatif des gènes codant les protéines cibles des nouvelles thérapies ciblées du cancer. / A genome-wide analysis of mutation, gene expression and alternative pre-mRNA splicing was performed in colorectal normal mucosa, adenoma and adenocarcinoma biopsy samples in order to look for some alterations that could characterize the stepwise “colorectal normal mucosa-adenoma-adenocarcinoma” transition. It was conducted through different microarray-based experiments. Alterations specific for either adenomas or adenocarcinomas were identified. Nevertheless, most deregulated genes in adenocarcinomas were shared between adenomas and adenocarcinomas, in agreement with the notion that adenomas are precursor lesions for adenocarcinomas. Adenomas may have different outcomes, depending on environment, some evolving towards cancer, while others could be prone to disappearance. Pathway enrichment in colorectal lesions and classification of colorectal lesions were investigated. A 40-gene set was identified as a gene expression signature that could help predicting patients, at time of adenoma ablation, with a risk for developing colorectal cancer. Splicing profiles were also identified in colorectal lesions, suggesting that alternative splicing may play a major role in cancer outcome. Finally, a custom microarray was designed with the aim to predict the response of patients before treatment. This custom microarray makes it possible to analyze transcript structure and levels for genes involved in the response to targeted anticancer therapies.

Philosophiae Doctor - PhD / Retinoblastoma binding protein 6 (RBBP6) is a 250 kDa protein that is implicated in mRNA processing and ubiquitination functions and has been shown to be highly up-regulated in a number of cancers. In humans and mice,RBBP6 interacts with both tumour suppressors p53 and pRb, suggesting that it is involved in regulation of transcription, induction of apoptosis and cell cycle control. Knock-out of an RBBP6 homologue PACT resulted in p53 dependent cell cycle arrest and apoptosis. Although the biological functions of RBBP6 remain largely unclear, it is possible that its functions are mediated through interaction with other cellular proteins. Since it is possible to unveil novel functions of a target protein through identifying its interacting protein partners,this study aims to further characterize the functions of RBBP6 through identifying novel protein interacting partners using a yeast 2-hybrid screen.In order to identify interaction partners of RBBP6, two well characterized domains of RBBP6, the N-terminal ubiquitin-like DWNN domain and RING finger domain, were used as baits in a yeast 2-hybrid screen of a human testis cDNA library. Putative interactors were verified using in vitro and in vivo immunoprecipitation assays. The RING finger domain was shown to interact with transcriptional factors Y-Box binding protein 1 (YB-1) and zinc finger and BTB containing protein 38 (zBTB38), resulting in their ubiquitination. In the case of YB-1 ubiquitination was correlated with a decrease in the intra-cellular levels of YB-1, suggesting that ubiquitination leads to degradation in the proteosome. The DWNN domain was shown to interact with a splicing associated small nuclear ribonucleoprotein polypeptide G (snRPG) and heat shock protein 70 (Hsp70).The results of this work suggest that, at least in the case of YB-1 and zBTB38,RBBP6 plays a role in the regulation of gene expression by ubiquitination of transcription factors, causing them to be degraded in the proteosome. The study provides further evidence of RBBP6’s involvement in mRNA splicing through its interaction with snRPG. The interaction with Hsp70 suggests a possible role in protein quality control similar to that played by other E3 ligases such as Parkin and CHIP.

Philosophiae Doctor - PhD / Retinoblastoma binding protein 6 (RBBP6) is a 250 kDa protein that is implicated in mRNA processing and ubiquitination functions and has been shown to be highly up-regulated in a number of cancers. In humans and mice, RBBP6 interacts with both tumour suppressors p53 and pRb, suggesting that it is involved in regulation of transcription, induction of apoptosis and cell cycle control. Knock-out of an RBBP6 homologue PACT resulted in p53 dependent cell cycle arrest and apoptosis. Although the biological functions of RBBP6 remain largely unclear, it is possible that its functions are mediated through interaction with other cellular proteins. Since it is possible to unveil novel functions of a target protein through identifying its interacting protein partners, this study aims to further characterize the functions of RBBP6 through identifying novel protein interacting partners using a yeast 2-hybrid screen. In order to identify interaction partners of RBBP6, two well characterized domains of RBBP6, the N-terminal ubiquitin-like DWNN domain and RING finger domain, were used as baits in a yeast 2-hybrid screen of a human testis cDNA library. Putative interactors were verified using in vitro and in vivo immunoprecipitation assays. The RING finger domain was shown to interact with transcriptional factors V-Box binding protein 1 (YB-1) and zinc finger and BTB containing protein 38 (zBTB38), resulting in their ubiquitination. In the case of YB-1 ubiquitination was correlated with a decrease in the intra-cellular levels of YB-1, suggesting that ubiquitination leads to degradation in the proteosome. The DWNN domain was shown to interact with a splicing associated small nuclear ribonucleoprotein polypeptide G (snRPG) and heat shock protein 70 (Hsp70). The results of this work suggest that, at least in the case of YB-1 and zBTB38, RBBP6 plays a role in the regulation of gene expression by ubiquitination of transcription factors, causing them to be degraded in the proteosome. The study provides further evidence of RBBP6's involvement in mRNA splicing through its interaction with snRPG. The interaction with Hsp70 suggests a possible role in protein quality control similar to that played by other E3 ligases such as Parkin and CHIP.

Pre-mRNA splicing is a process in which introns are removed from eukaryotic transcripts and exons are ligated together. Splicing is catalyzed by spliceosome, a large ribonucleoprotein complex composed of five small nuclear RNAs and more than 100 additional proteins, which recognizes 5' splice site, branch point site and 3' splice site and performs two transesterification reactions to produce mRNA molecules. 5' splice site is recognized by U1 snRNP and U2 auxiliary factor (U2AF) is involved in branch point and 3' splice site recognition in the early splicing complex. There is some evidence of splice sites cooperation during intron recognition in vitro but little is known about the situation in vivo. Using Fluorescence resonance energy transfer (FRET) and RNA immunoprecipitation (RIP) methods, we have investigated the early stages of spliceosome assembly. We have employed splicing reporters based on -globin gene and MS2 stem loops to detect interactions of proteins on RNA molecule directly in the cell nucleus. Results of FRET indicate that intact 5' splice site is required for U2AF35 interaction with 3' splice site and that U1C recruitment to 5' splice site is partially limited upon 3' splice site mutation. We have also confirmed by RIP that U2 snRNP association with pre-mRNA molecule requires presence of 5'...

Alternative splicing is one of the mechanisms how to regulate gene expression. Under different conditions, different mRNAs encoding proteins with different function, localization or stability can be made from one cellular transcript. The human hSlu7 protein affects the alternative splicing of some genes through alternative 3'splice site (3'SS) selection. Although it was thought that alternative splicing is absent from Saccharomyces cerevisiae, recent results argue against such conclusion. We therefore decided to characterize the function of the yeast Slu7 protein, which participates in the second step of splicing and is closely associated with the 3'SS selection. We focused on a highly conserved uncharacterized motif in the essential part of the Slu7 protein named the RED motif. Mutations in this motif caused second step splicing defects with some substrates and altered the alternative 3'SS usage ratio of some splicing constructs. Our results implicate a role for the RED motif in selecting proper 3'splice sites, especially the distal ones. Genetic interactions of slu7 mutations with PRP22 and PRP45 mutant alelles add to the intricate interaction network of splicing factors and suggest a possible role of Slu7p in facilitating the Prp22p association with the spliceosome.

Saccharomyces cerevisiae is an intron-poor organism with introns present in only 5% of its genes. The most prominent group of intron-containing genes are ribosomal protein (RP) genes. They are highly expressed and most of them are present as two paralogs. Parenteau et al. described the existence of intron- dependent intergenic regulatory circuits controlling expression ratios of RP paralogs. In this project, we did not confirm the regulation in 6 out of 7 tested regulatory circuits. We validated the regulation between RPL22 paralogs. We further showed that Rpl22 protein blocks the pre-mRNA splicing of both paralogs, with RPL22B paralog being more sensitive. Rpl22 protein binds to the stem-loop of RPL22B intron - disruption of the binding domain of Rpl22 proteins leads to loss of interaction. Moreover, the regulation seems to be working the same way in yeast Kluyveromyces lactis, which has only a single RPL22 copy. Overall, these results lead to better understanding of intergenic regulation, which adjusts the expression ratio between functionally different RPL22 paralogs. Key words introns, ribosomal protein genes, Rpl22, RPL22 paralogs, pre-mRNA splicing, Saccharomyces cerevisiae

Our laboratory delineated a role for alternative pre-mRNA splicing (AS) in triple negative breast cancer (TNBC). We found the translational regulator cytosolic polyadenylation element binding protein 2 (CPEB2) which has two isoforms, CPEB2A and CPEB2B, is alternatively spliced during acquisition of anoikis resistance (AnR) and metastasis. The splicing event which determines the CPEB2 isoform is via inclusion/ exclusion of exon four in the mature mRNA transcript. The loss of CPEB2A with a concomitant increase in CPEB2B is required for TNBC cells to metastasize in vivo. We examined RNAseq profiles of TNBC cells which had CPEB2 isoforms specifically downregulated to examine the mechanism by which CPEB2 isoforms mediate opposing effects on cancer-related phenotypes. Downregulation of the CPEB2B isoform inhibited pathways driving the epithelial-to-mesenchymal transition (EMT) and hypoxic response, whereas downregulation of the CPEB2A isoform did not have this effect. Specifically, CPEB2B functioned as a translational activator of TWIST1 and HIF1a. Functional studies showed that specific downregulation of either HIF1α or TWIST1 inhibited the ability of CPEB2B to induce AnR and drive metastasis. The mechanism governing inclusion/ exclusion of exon 4 was determined to be serine/ arginine-rich splicing factor 3 (SRSF3). Binding of SRSF3 to a consensus sequence within CPEB2 exon 4 promoted its inclusion in the mature mRNA, and mutation of this sequence abolished association of SRSF3 with exon 4. SRSF3 expression was upregulated in TNBC cells upon acquisition of AnR correlating with a reduction in the CPEB2A/B ratio. Importantly, downregulation of SRSF3 by siRNA in these cells induced the exclusion of exon 4. Downregulation of SRSF3 also reversed the CPEB2A/B ratio in a wild-type CPEB2 exon 4 minigene construct, but not a mutant CPEB2 minigene with the SRSF3 RNA cis-element ablated. Physiologic studies demonstrated SRSF3 downregulation ablated AnR in TNBC cells, and was “rescued” by ectopic expression of CPEB2B. Importantly, biostatistical analysis of The Cancer Genome Atlas database showed a positive relationship between alterations in SRSF3 expression and lower overall survival in TNBC. Overall, this study demonstrates that SRSF3 modulates CPEB2 AS to induce the expression of the CPEB2B isoform that drives TNBC phenotypes correlating with aggressive human breast cancer.

<p>ASF/SF2 is an essential splicing factor, required for constitutive splicing, and functioning as a regulator of alternative splicing. ASF/SF2 is modular in structure and contains two amino-terminal RNA binding domains (RBD1 and RBD2), and a carboxy-terminal RS domain. The results from my studies show that the different activities of ASF/SF2 as a regulator of alternative 5’ and 3’ splice site selection can be attributed to distinct domains of ASF/SF2.</p><p>I show that ASF/SF2-RBD2 is both necessary and sufficient to reproduce the splicing repressor function of ASF/SF2. A SWQDLKD motif was shown to be essential for the splicing repressor activity of ASF/SF2. In conclusion, this study demonstrated that ASF/SF2 encodes for distinct domains responsible for its function as a splicing enhancer (the RS domain) or a splicing repressor (the RBD2) protein. Using a model transcript containing two competing 3’ splice sites it was further demonstrated that the activity of ASF/SF2 as a regulator of alternative 3’ splice site selection was directional: i.e. resulting in RS or RBD1 mediated activation of upstream 3’ splice site selection while simultaneously causing an RBD2 mediated repression of downstream 3’ splice site usage.</p><p>In alternative 5’ splice site selection, the RBD2 alone was sufficient to reproduce the activity of the full-length protein as an inducer of proximal 5’ splice site usage, while RBD1 had the opposite effect and induced distal 5’ splice site selection. The conserved SWQDLKD motif and the RNP-1 type RNA recognition motif in ASF/SF2-RBD2 were both essential for this induction. The activity of the ASF/SF2-RBD2 domain as a regulator of alternative 5’ splice site was shown to correlate with the RNA binding capacity of the domain.</p><p>Collectively, my results suggest that the RBD2 domain in ASF/SF2 plays the most decisive role in the alternative 5’ and 3’ splice site regulatory activities of ASF/SF2.</p>

ASF/SF2 is an essential splicing factor, required for constitutive splicing, and functioning as a regulator of alternative splicing. ASF/SF2 is modular in structure and contains two amino-terminal RNA binding domains (RBD1 and RBD2), and a carboxy-terminal RS domain. The results from my studies show that the different activities of ASF/SF2 as a regulator of alternative 5’ and 3’ splice site selection can be attributed to distinct domains of ASF/SF2. I show that ASF/SF2-RBD2 is both necessary and sufficient to reproduce the splicing repressor function of ASF/SF2. A SWQDLKD motif was shown to be essential for the splicing repressor activity of ASF/SF2. In conclusion, this study demonstrated that ASF/SF2 encodes for distinct domains responsible for its function as a splicing enhancer (the RS domain) or a splicing repressor (the RBD2) protein. Using a model transcript containing two competing 3’ splice sites it was further demonstrated that the activity of ASF/SF2 as a regulator of alternative 3’ splice site selection was directional: i.e. resulting in RS or RBD1 mediated activation of upstream 3’ splice site selection while simultaneously causing an RBD2 mediated repression of downstream 3’ splice site usage. In alternative 5’ splice site selection, the RBD2 alone was sufficient to reproduce the activity of the full-length protein as an inducer of proximal 5’ splice site usage, while RBD1 had the opposite effect and induced distal 5’ splice site selection. The conserved SWQDLKD motif and the RNP-1 type RNA recognition motif in ASF/SF2-RBD2 were both essential for this induction. The activity of the ASF/SF2-RBD2 domain as a regulator of alternative 5’ splice site was shown to correlate with the RNA binding capacity of the domain. Collectively, my results suggest that the RBD2 domain in ASF/SF2 plays the most decisive role in the alternative 5’ and 3’ splice site regulatory activities of ASF/SF2.

The presence of introns in all the eukaryotic genomes identified so far underscores the fundamental and ubiquitous role of pre-mRNA splicing. The spliceosomal machinery, comprised of five small nuclear RNAs and several protein factors, catalyzes the two-transesterification reactions of splicing with precision and consistency. Through a complex network of protein-protein and RNA-protein interactions it ensures the removal of the intron and ligation of the flanking exons to yield the mature mRNA. Prpl7 is a splicing factor that functions at the second-step of splicing (Vijayraghavan et all, 1989). Null alleles of prpl7 are viable at 23°C but die at temperatures above 33°C (Jones et al.9 1995). Besides its functions in pre-mRNA splicing, mutants in PRP17ICDC40 were independently shown to affect cell-cycle progression, particularly the Gl/S and G2/M transitions (Chawla et a/., 2003). In this study, we have attempted a further characterization of Prpl7 to analyze both its role in pre-mRNA splicing and in cell-cycle progression with an aim to decipher underlying reasons for the interlinking of these two cellular processes. Different experimental approaches were adopted to achieve this goal. Global gene-expression profiling provided an overview of all the transcripts affected in a prpl 7 mutant and allowed its comparison with mutants of other splicing factors. This exercise aided in identification of both pre-mRNA splicing and cell-cycle related effects of Prpl7. Biochemical analysis of the Prpl7 spliceosomal associations have provided further clarity on the part played by Prpl7 in pre-mRNA splicing. A genome-wide two-Hybrid screen for interacting partners of Prpl7 was undertaken and uncovered two Likely interacting partners of Prpl7. Global expression profiling of splicing mutants Pleiotropic phenotypes observed in mutants of prpl 7 and few other splicing factors have been speculated to arise from either the multi functionality of the factor or more likely due to a specific requirement of the factor in splicing of a select subset of transcripts, that encode proteins essential to the affected cellular pathway. These observations raise questions about the ubiquitous requirement of factors in pre-mRNA splicing. To understand these aspects of splicing, we studied the effects of splicing factor mutants on a genome-wide scale. Using splicing-sensitive DNA microarrays imprinted with all yeast ORFs and in addition, independent spots for a majority of the intron sequences, we analyzed the global expression changes triggered by the inactivation of temperature-sensitive mutations in PRP17 or PRP22. Experiments with prp2-l mutant strain detect, as expected, an increase in pre-mRNA levels at the intron spots and further demonstrated that the ORF spots detect a decrease in mRNA levels in these DNA microarrays. These results established the DNA micro arrays as tools for the analysis of splicing on a global scale. The temporal alterations in transcript profiles in prpl 7 and prp22 mutants, as compared to the wild type, revealed both shared and unique effects of these factors on clusters of intron-containing transcripts. Such differential effects, on intron-containing transcripts, amongst the splicing mutants implicate specialized roles for each of these factors. Through analysis of the set of intron-containing transcripts affected in prpl7Δ cells, we infer those attributes of these pre-mRNA substrates, which predispose a need for Prpl 7. We find that splicing of introns longer than 200nts has a stronger dependence on Prpl7. The distance between consensus intron elements- the branch-nucleotide and the 3'splice-site (B), also imposes a requirement for Prpl7. Introns with a 13nts or lesser distance between these elements are spliced even in the absence of Prpl 7, both in vivo and in vitro. The 5'splice-site to branch-nucleotide distance (A) also influences the need for Prpl7. Most introns with a A/B ratio of less than 2 undergo Prpl7 independent splicing in vivo. Intron-containing genes that could be responsible for the pleiotropic phenotypes of prpl7 were also identified through the global splicing analysis. These included splicing targets that act at the Gl-S phase such as ANC1/TAF14, TMD4, PHO85 and those at the G2-M phase of the cell-cycle; TUB], TUB3, GIM5, MOBl UBC9. Recently, a different study implicates ANC1ITAF14 as the intron-containing gene responsible for the cell-cycle phenotype associated with prpl7 (Dahan and Kupiec, 2004). Our global analysis of all intron-containing transcripts with compromised expression in prpl7A cells identify, in addition, PHO85 as a possible regulator underlying cell-cycle effects in this mutant. Pho85 is a cyclin-dependent kinase that functions at both the Gl/S and M/Gl phases of the division cycle (Moffate* al., 2000). Synergistic growth defects in double mutants of prpl7 and pho85 have uncovered a novel role for Prpl7 in bud morphogenesis. Our micro array data also reveals compromised expression levels for several key intronless cell-cycle rregulatory genes indicating a possible splicing-independent role for Prpl7 in the cell-cycle. Examples of such transcripts are: the Gl cyclins CLN1, CLN2 and CLN3; CDC6, required for assembly of the pre-replication complex at sites of replication origin; and the cell-cycle regulatory transcription factors: SWI5 and ACE2. The global analysis has therefore enabled, for the first time, a characterization of the splicing substrate specificity of Prpl7 and has also uncovered the effects of this protein on gene expression during cell-cycle progression (Fig. V.I A). Spliceosomal interactions of Prpl7 To understand the function and associations of Prpl7 in the spliceosome, we have examined its snRNP interactions and determined the time point of its coalescence on assembling spliceosomes. A functional epitope tagged-Prpl7 was created using the polyoma middle T-antigen and the poly-HIS tags (Stevens et aln 1999). Through immunoprecipitation analyses performed with splicing extracts, from this strain, we find Prpl7 to associate with three spliceosomal snRNPs- U2, U5 and U6, implicating an interaction with active spliceosomes or post-splicing complexes. Specific biochemical depletion of any one of these snRNAs, through oligo-directed RNaseH cleavage, did not have a drastic effect on the association of Prpl7 with the other two snRNAs. To decipher the point at which Prp 17 joins the assembling spliceosomes, we examined the presence of Prp 17 in in vitro assembled complexes generated under various conditions. The conditions adopted were designed to stall and enrich for •assembly intermediates. A co-immunoprecipitation of the input precursor RNA and reaction intermediates revealed an early association of Prp 17 with the assembling Spliceosome prior to its catalytic activation. This association occurred in the A2-1 complex, which contains the U4/U6.U5 tri-snRNP along with the Ul and U2snRNPs. Prpl7 was found to associate with all subsequent complexes until the completion of catalytic transesterification reactions and possibly continue with the spliced-out introns complex (Fig. V.1B). Identification of two novel interacting partners of Prpl7 from a genome-wide two-hybrid screen Although several genetic interacting partners of PRP17 are known, none display a direct physical association with Prpl7. Knowledge of the proteins that Prpl7 interacts with can further the functional characterization of this protein and aid in deciphering its link to cell-cycle progression. A genome-wide screen for interacting partners using Prpl7 as bait was carried out in a two-hybrid system with a yeast genomic DNA-B42 activation-domain library (Gyuris et al., 1993). Through this screen we identified two interacting partners of Prpl 7- YOL078W, an essential gene and SGML The domain in the 1176 amino acid YOL078W protein responsible for interaction with Prpl7 was mapped to a 225 amino acid segment in the C-terminai region of this protein. The N-terminal region of the protein appears to exert a negative effect on the interaction with Prpl7. While YOL078w does not have any apparent role in pre-mRNA splicing, a majority of the cells arrest with small buds indicating a late Gl or early S phase arrest upon transcriptional shut-down of YOL078W. YOL078W has been independently characterized as AVOl, a component of the TOR complex, involved in nutrient sensing and cell size regulation (Loewith et al, 2002). Other reports show it tto be a component of a complex that interacts with Ceglp, a nuclear protein involved in mRNA capping (Gavin et al, 2002). We hypothesize that Prpl7 and Avol may exist in a dynamic nucleocytoplasmic complex possibly functioning in either cell-cycle regulation, RNA processing or both. Through this study we have Established the use of splicing-sensitive microarrays as tools for the characterization of pre-mRNA splicing factors. Simultaneous assessment of the effects on other cellular pathways was accomplished through expression profiling of all the intron-containing and intronless genes. Deciphered the differential dependence of pre-mRNA substrates on spliceosome factors at a global scale. Predicted the substrate-specificity of the second-step splicing factor, Prpl7, and verified some of these predictions in vitro. Gathered evidence for a possible splicing-independent effect of Prpl7 on the cell division cycle. Uncovered a novel function of Prpl7 in bud morphogenesis, as deduced from its synergistic genetic interaction with PHO85. Identified U2, U5 and U6 snRNPs as interacting partners of Prpl7 in both xtracts and in in vitro splicing reactions. Determined the point of coalescence of Prpl7 during spliceosome assembly to be at an early assembly stage soon after the entry of U4/U6.U5 tri-snRNP and prior to catalytic activation. Demonstrated continued Prpl7 association with the spliceosome beyond the completion of the splicing reactions. Identified Avolp and Sgmlp as novel interacting partners of Prpl7 through a genome-wide two-hybrid screen.