Ubiquitination complex CRL4 (Cullin Ring Ligase) attracts a lot of attention due to its involvement in physiological and pathological processes, especially in the development of cancer. Cullin4 a/b proteins are reported to serve as oncoproteins in various malignancies. Due to their role in the regulation of cancer drugs targeting CRL4 have been identified, including thalidomide and its derivatives inhibiting one of the substrate receptors of the complex, the Cereblon protein. The adapter protein within the CRL4 complex - DDB1, which is involved i.a. in DNA repair, also has a role in cancer. However, the mechanism of this function has not yet been fully elucidated. The subject of this master thesis was to study the effects of elimination and suppression of CRL4 complex functions in the prostate cancer cell line LNCaP. Significantly variable changes in cell proliferation and migration have been observed if the complex functions were affected by thalidomide. The creation of the LNCaP cell line with conditionally suppressed DDB1 function was used to study tumor dynamics in a mouse model. Results show that suppression of DDB1 function has an inhibitory effect on tumor cell proliferation but increases their ability to invade adjacent tissues. Complete deletion of the DDB1 gene in the LNCaP cell line...

La dégradation sélective des protéines par un mécanisme ubiquitine et protéasome 26S dépendant est conservée chez tous les eucaryotes. Les E3 ubiquitine ligases sont les derniers acteurs de la cascade d’ubiquitinylation et sont responsables de la sélection spécifique des protéines cibles pour leur dégradation. Les enzymes E3 de type CRL4 forment des complexes multiprotéiques dont les récepteurs de substrats sont appelés DCAF pour DDB1-CUL4 Associated Factor. Les études réalisées chez les mammifères et les levures ont permis d’identifier une signature spécifique des DCAF : le motif WDxR à la fin d’un domaine WD40. L’analyse bioinformatique a montré l’existence de plus de 85 DCAF potentielles chez Arabidopsis thaliana. Parmi ces récepteurs, nous nous sommes intéressés à une famille multigénique appelée MSI (Multicopy Suppressor of IRA1). Des études précédentes ont permis de montrer que MSI1 et MSI4 participaient chacune à un complexe CRL4 différent capable d’interagir fonctionnellement avec un complexe PRC2 (Polycomb Repressive Complex 2) pour moduler une régulation épigénétique.Les résultats obtenus et décrits dans ce manuscrit mettent en évidence une interaction physique entre les protéines MSI (2 ou 3) et DDB1a suggérant l’existence d’un complexe multimérique incluant CUL4. L’interrogation des données publiques confrontées à nos données expérimentales, nous a permis d’établir que l’expression des deux gènes était régulée de manière cycle cellulaire dépendante. De plus, un mutant perte de fonction msi3 présente un phénotype de retard de croissance suggérant une fonction de contrôle du cycle cellulaire. D’autre part, leur capacité d’interagir avec des régulateurs chromatiniens permet d’envisager une régulation par voie épigénétique. Toutefois, le rôle exact de ces protéines reste à déterminer. / Selective protein degradation by the UPS (Ubiquitin Proteasome System) is highly conserved in all eukaryotes. The E3 ubiquitin ligases are the last actors in the ubiquitylation pathway and target specifically the proteins for degradation. CRL4 E3 ligases form multiprotein complexes where the substrate receptors are called DCAFs for DDB1-CUL4 Associated Factor. Studies made in mammals and yeast allowed to highlight a characteristic signature for the DCAFs: the WDxR motif at the end of a WD40 domain. Bioinformatic studies could identify around 85 potential DCAFs in Arabidopsis thaliana. Among those receptors, we were interested in a small multigenic family called MSIs (Multicopy Suppressor of IRA1). Previous studies showed that MSI1 and MSI4 belong to different CRL4 complexes functionally connected to a PRC2 complex (Polycomb Repressive Complex 2). The results obtained and described in this manuscript highlight a physical interaction between MSI (2 or 3) and DDB1a suggesting the existence of a multimeric complex including CUL4. Furthermore, bioinformatic as well as experimental data, allowed us to establish that MSI2 and MSI3 gene expression are cell cycle regulated. Moreover, phenotypic analysis of an msi3 loss of function mutant showed a delayed growth implying a function as cell cycle regulator. On the other hand, the ability of MSI3 to interact with chromatin regulators points to an epigenetic regulatory pathway. However, the exact function of these proteins remains to be determined.

Merlin is a tumour suppressor protein that is frequently mutated or downregulated in cancer. Biallelic Merlin inactivation is causative of tumour formation, including schwannoma, meningioma and ependymoma. These tumours can occur sporadically or as part of the genetic condition Neurofibromatosis type 2 (NF2) and cause significant morbidity. The current treatment options are restricted to surgery and radiotherapy, which are invasive and may cause further tumour development. The activity of both the E3 ubiquitin ligase complex Cullin 4 really interesting new gene (RING) E3 ubiquitin ligase- DNA damage binding protein (DDB1) and Cullin 4 associated factor 1 (CRL4-DCAF1) and Kinase suppressor of RAS 1 (KSR1) have been shown to be upregulated in schwannoma to drive tumour growth. KSR1 has also been shown to interact with components of the CRL4-DCAF1 complex. We investigated the expression, interaction and therapeutic potential of targeting these proteins in Merlin deficient schwannoma and meningioma using a primary human cell model and relevant cell lines. We found that DCAF1 and KSR1 protein were overexpressed in schwannoma and meningioma and confirmed that targeting both DCAF1 and KSR1 in meningioma had additive effects on proliferation. We also identified that CRL4-DCAF1 facilitates KSR1 dependent RAF/Mitogen-activated protein kinase (MAPK)/ Extracellular signal regulated kinase (ERK) kinase (MEK)/ERK pathway activity. We showed MLN3651, a neddylation inhibitor that targets ubiquitin ligase activity, reduced proliferation and activated apoptosis in Merlin-deficient tumours. We also showed that Merlin-positive tumours were less sensitive to MLN3651 than Merlin-deficient tumours; therefore, MLN3651 sensitivity may be CRL4-DCAF1-dependent. Finally, combination of MLN3651 and the MEK1/2 inhibitor AZD6244 had additive effects, particularly in meningioma. Combinatorial therapy activated the Hippo pathway, inhibited RAF/MEK/ERK pathway activity and proliferation demonstrating that targeting the activity and downstream pathways of both DCAF1 and KSR1 represents an attractive novel therapeutic strategy in Merlin-deficient tumours.

The E3 ubiquitin ligase \(CRL4^{Cdt2}\) targets proteins for destruction during DNA replication and following DNA damage (Havens and Walter, 2011). Its substrates contain "PIP degrons" that mediate substrate binding to the processivity factor PCNA at replication forks and damage sites. The resulting PCNA-PIP degron complex forms a docking site for \(CRL4^{Cdt2}\), which ubiquitylates the substrate on chromatin. Several \(CRL4^{Cdt2}\) substrates are known, including Cdt1, multiple CDK inhibitors, Drosophila E2f1, human Set8, S. pombe Spd1, and C. elegans \(Pol\eta\) (Havens and Walter, 2011). An emerging theme is that \(CRL4^{Cdt2}\) targets proteins whose presence in S phase is toxic. Here, I used Xenopus egg extract to characterize a new \(CRL4^{Cdt2}\) substrate, thymine DNA glycosylase (TDG). TDG is a base excision repair protein that targets G-U and G-T mispairs, which arise from cytosine and 5-methylcytosine deamination (Cortazar et al., 2007). Thus, TDG may function in epigenetic gene regulation via DNA demethylation, in addition to its canonical DNA repair function. A yet unknown E3 ubiquitin ligase triggers TDG destruction during S phase (Hardeland et al., 2007). Understanding TDG proteolysis in S phase is relevant to the regulation of DNA replication, DNA repair, and epigenetic control of gene expression. I discovered that TDG contains a variant of the "PIP degron" consensus and that TDG is ubiquitylated and destroyed in a PCNA-, Cdt2-, and degron-specific manner during DNA repair and DNA replication in Xenopus egg extract. I further characterized what features of TDG contribute to its proteolysis. Interestingly, I could not identify any defects during DNA replication or during Xenopus embryonic development in response to a non-degradable form of TDG. Additionally, I examined how interactions between \(CRL4^{Cdt2}\) and multiple subunits of the PCNA homotrimer contribute to \(CRL4^{Cdt2}\) function. In a popular model, PCNA functions as a "tool belt" on DNA, binding three separate proteins through its individual subunits to facilitate rapid exchange of DNA replication and repair proteins as they are needed on DNA. To address this model, I generated a single chain polypeptide with three PCNA subunits connected through flexible linker sequences. I used this tool to determine how multiple PCNA subunits contribute to \(CRL4^{Cdt2}\) function. I found that a single wildtype subunit is sufficient for modest destruction of the \(CRL4^{Cdt2}\) substrate Cdt1, but complete Cdt1 destruction requires two separate wildtype subunits. Additionally, a single subunit was sufficient for leading strand elongation, challenging the "tool belt" model during DNA replication. I also discuss implications and future use of the single-chain PCNA.