Fat facets in Mouse (FAM) or mUSP9x is a deubiquitylating enzyme of the USP class. Knockdown of FAM protein levels in mouse pre-implantation embryos by antisense oligonucleotides is known to prevent embryos from progressing to the blastocyst stage indicating an important role for FAM in early mammalian development. In mammals, the Fam gene is located on the X-chromosome. In mice, the Y homologue, Dffry or usp9y, is expressed exclusively in the testes and maps to the Sxrb deletion (Brown et al., 1998). Sxrb is associated with an early post-natal blockage of spermatogonial proliferation and differentiation leading to absence of germ cells (Bishop et al., 1988; Mardon et al., 1989). The human Y homologue of Fam is closely associated with oligozoospermia (Sargent et al., 1999; Sun et al., 1999) and the human X homologue has been linked to the failure of oocytes to pass through the first meitoc prophase in Turner syndrome (Cockwell et al., 1991; Speed, 1986) Despite these associations, the substrates and precise role of Fam and its homologues in these processes have not yet been defined. Due to the complex nature of Fam expression and the lack of data tying FAM to specific cellular functions, much attention has been paid in identifying interacting partners and cellular targets of FAM activity to aid in the definition of its role in the cell and development. Three common molecular biology techniques were applied here in an attempt to further characterise known interactions of FAM, including interactions with the cell adhesion molecule β-catenin and the protein trafficking pathway proteins epsin-1 and itch. The aim of these investigations was to generate FAM mutants that could abolish individual interactions, enabling investigation of individual interactions in cellular function and development. These experiments failed to identify the amino acids of FAM that were critical for its interactions with β-catenin, epsin-1, or itch. Experiments aimed at characterising a novel ubiquitin-like domain located in the N-terminal half of the FAM protein, did however identify novel interactions of FAM with the three Golgi associated adaptor proteins GGA1, GGA2, and GGA3. Further investigations prompted by this interaction, examined the role of FAM in the trafficking of proteins from the Golgi apparatus. Cellular FAM protein levels were altered either by exogenous expression of FAM protein or knockdown of endogenous FAM using FAM specific shRNA triggers. The cellular protein levels and extent of post-translational modification of eleven lysosomal proteins were monitored in each case. It was found that increased FAM protein levels resulted in decreased cellular protein levels of five of the eleven lysosomal proteins studied. In contrast, a reduction in FAM protein levels was found to result in an increase in the cellular protein levels of eight of the eleven lysosomal proteins. This study provides the first evidence of a deubiquitylating enzyme that is able to interact with the GGA proteins. It is also the first to describe a deubiquitylating enzyme that can affect the biosynthesis of lysosomal proteins and provides valuable new insight into the cellular function of FAM/USP9X. / Thesis (Ph.D.) -- University of Adelaide, School of Molecular and Biomedical Sciences, 2008

Fat facets in Mouse (FAM) or mUSP9x is a deubiquitylating enzyme of the USP class. Knockdown of FAM protein levels in mouse pre-implantation embryos by antisense oligonucleotides is known to prevent embryos from progressing to the blastocyst stage indicating an important role for FAM in early mammalian development. In mammals, the Fam gene is located on the X-chromosome. In mice, the Y homologue, Dffry or usp9y, is expressed exclusively in the testes and maps to the Sxrb deletion (Brown et al., 1998). Sxrb is associated with an early post-natal blockage of spermatogonial proliferation and differentiation leading to absence of germ cells (Bishop et al., 1988; Mardon et al., 1989). The human Y homologue of Fam is closely associated with oligozoospermia (Sargent et al., 1999; Sun et al., 1999) and the human X homologue has been linked to the failure of oocytes to pass through the first meitoc prophase in Turner syndrome (Cockwell et al., 1991; Speed, 1986) Despite these associations, the substrates and precise role of Fam and its homologues in these processes have not yet been defined. Due to the complex nature of Fam expression and the lack of data tying FAM to specific cellular functions, much attention has been paid in identifying interacting partners and cellular targets of FAM activity to aid in the definition of its role in the cell and development. Three common molecular biology techniques were applied here in an attempt to further characterise known interactions of FAM, including interactions with the cell adhesion molecule β-catenin and the protein trafficking pathway proteins epsin-1 and itch. The aim of these investigations was to generate FAM mutants that could abolish individual interactions, enabling investigation of individual interactions in cellular function and development. These experiments failed to identify the amino acids of FAM that were critical for its interactions with β-catenin, epsin-1, or itch. Experiments aimed at characterising a novel ubiquitin-like domain located in the N-terminal half of the FAM protein, did however identify novel interactions of FAM with the three Golgi associated adaptor proteins GGA1, GGA2, and GGA3. Further investigations prompted by this interaction, examined the role of FAM in the trafficking of proteins from the Golgi apparatus. Cellular FAM protein levels were altered either by exogenous expression of FAM protein or knockdown of endogenous FAM using FAM specific shRNA triggers. The cellular protein levels and extent of post-translational modification of eleven lysosomal proteins were monitored in each case. It was found that increased FAM protein levels resulted in decreased cellular protein levels of five of the eleven lysosomal proteins studied. In contrast, a reduction in FAM protein levels was found to result in an increase in the cellular protein levels of eight of the eleven lysosomal proteins. This study provides the first evidence of a deubiquitylating enzyme that is able to interact with the GGA proteins. It is also the first to describe a deubiquitylating enzyme that can affect the biosynthesis of lysosomal proteins and provides valuable new insight into the cellular function of FAM/USP9X. / Thesis (Ph.D.) -- University of Adelaide, School of Molecular and Biomedical Sciences, 2008

Ubiquitylation is an important post-translational modification and can be reversed by the action of deubiquitinating (DUB) enzymes. The ubiquitylation and deubiquitylation of target proteins are significant in terms of regulating cellular events such as protein degradation, signal transduction, vesicle trafficking, DNA repair and apoptosis. Chromosomal band 17q23 is frequently amplified in breast cancers and harbors a predicted ubiquitin specific protease gene, USP32 (ubiquitin specific protease 32). Given its potential role in breast cancer, we aimed to characterize USP32 for its potential DUB activity. Bioinformatic analysis of USP32 and known yeast and mouse DUBs suggested presence of Cys-His domains which are common in active DUBs of the USP superfamily. Our in vivo and in vitro DUB activity assays revealed that USP32 was indeed an active deubiquitinating enzyme. To investigate its substrate specificity and kinetic properties, USP32 was expressed in insect cell culture to be isolated and purified. Using isolated USP32 protein, diubiquitin assay was performed with all seven types of diubiquitin (K6, K11, K27, K29, K33, K48 and K63) as well as linear diubiquitin. Results showed that USP32 was able to cleave all seven types of ubiquitin linkages with higher cleavage efficiency for K6, K11, K48 and K63-linked diubiquitin. Moreover, kinetic parameters, Km, kcat and kcat/ Km, suggested that full length protein had lower affinity for potential substrates and lower catalytic activity compared to the catalytic domain alone. These data suggested the importance of USP32 tertiary structure and possible role of other non DUB domains (e.g. EF hand domain) which may be regulated by an as of unknown mechanism in cells. Further investigations are underway to understand the functions of USP32 in cells and how it may contribute to breast tumorigenesis.

En dégradant la majorité des protéines cellulaires, le protéasome se positionne comme un régulateur clé du protéome, vis-à-vis duquel la plupart des tumeurs présentent une forte addiction en raison du débalancement protéique qui les caractérise. Quoique son inhibition se soit avérée être une bonne stratégie anticancer, elle est demeurée limitée aux cancers sanguins. Malheureusement, leur traitement devient tôt ou tard compromis par la résistance cellulaire. Raison pour laquelle l'élucidation du mécanisme de mort en jeu pourrait permettre de mieux cerner cette résistance, ce qui constituerait les fondements pour un traitement plus efficace. L'un des événements les plus spectaculaires et les plus précoces à se manifester dans le cadre de ce traitement, est la déubiquitination massive de l'histone H2A sur la lysine (K) 119. Une corrélation positive plutôt paradoxale entre cet événement, qui est associé à l'expression génique, et la sensibilité cellulaire à l'inhibition du protéasome, a été remarquée. Cela a mené à s'intéresser à sa signification biologique. Des cellules cancéreuses et primaires ont servi de systèmes d'étude protéomique par immunobuvardage et par immunofluorescence, pour analyser l'état de la chromatine et la distribution spatio-temporelle des histones durant l'inhibition du protéasome. Des inhibiteurs chimiques, des ARN interférents et des vecteurs d'expression ont été utilisés à cette fin. Un impressionnant phénomène survenant à la suite de l'inhibition du protéasome a été révélé. En effet, une baisse drastique du niveau d'histones sur la chromatine s'opère simultanément à la déubiquitination de H2A-ub (K119). Le protéasome étant inhibé, celles-ci, et possiblement les histones synthétisées en phase S, subiraient une translocation irréversible dans les nucléoles, et ce avant le déclenchement de l'apoptose. Ce phénomène est reproduit par divers inhibiteurs du protéasome et par siRNA, et il survient autant dans des cellules cancéreuses que primaires, mais pas dans les cellules résistantes, qui ne démontrent d'ailleurs pas de déubiquitination de H2A-ub (K119). Par ailleurs, il a été montré que la surexpression d'histones exogènes mène à leur translocation nucléolaire, et que la combinaison de l'inhibition du protéasome à cette surexpression pourrait être léthale. Quoique majoritairement préliminaires, les résultats révèleraient un surprenant mécanisme de régulation post-traductionnelle des histones endogènes, qui seraient séquestrées dans les nucléoles lorsqu'elles ne sont pas incorporées à la chromatine. En effet, l'inhibition du protéasome occasionne une importante perturbation de la chromatine pendant plusieurs heures. En raison de la cytotoxicité intrinsèque des histones libres et de leur abondance dans les cellules, celles-ci pourraient bien être à l'origine de la mort induite par l'inhibition du protéasome. Enfin, en sa qualité de senseur majeur de stress, le nucléole pourrait bien être le point de départ de la signalisation menant à la mort. / By degrading most of cellular proteins, the proteasome is positioned as a key regulator of the proteome, against which most tumors have a strong addiction, due to the protein imbalance that characterizes them. Although its inhibition has been shown to be a good anticancer strategy, it is still limited to blood cancers. Unfortunately, their treatment sooner or later becomes compromised by cellular resistance. This is why the elucidation of the mechanism of death involved could allow a better understanding of this resistance, which would in turn constitute the basis for a more effective treatment One of the most spectacular and early events to manifest during this treatment is the massive deubiquitylation of histone H2A on lysine (K) 119. A rather paradoxical positive correlation between this event, which is associated with gene expression, and cellular sensitivity to proteasome inhibition, has been noticed. This led to an interest in its biological significance. Cancer and primary cells have been used as systems for proteomic study by immunoblot and immunofluorescence, to analyze chromatin status and the spatio-temporal distribution of histones during proteasome inhibition. Chemical inhibitors, interfering RNAs and expression vectors have been used for this purpose. An impressive phenomenon occurring during the proteasome inhibition has been revealed. Indeed, a drastic drop in histones level on chromatin occurs simultaneously with the deubiquitylation of H2A-ub (K119). The proteasome being inhibited, these, and possibly histones synthesized in S phase, would undergo an irreversible translocation in the nucleoli before the onset of apoptosis. This phenomenon is replicated by various proteasome inhibitors and siRNA, and occurs in both cancer and primary cells, but not in resistant cells, which do not demonstrate deubiquitination of H2A-ub (K119). Furthermore, overexpression of exogenous histones has been shown to lead to their nucleolar translocation, and it is thought that the combination of proteasome inhibition with this overexpression could be lethal. Although mostly preliminary, the results would reveal a surprising mechanism of post-translational regulation of endogenous histones, which would be sequestered in nucleoli when not incorporated into chromatin. Indeed, inhibition of the proteasome causes a significant disruption of the chromatin for several hours. Due to the intrinsic cytotoxicity of free histones and to their great cellular abundance, these may well be the cause of the death induced by proteasome inhibition. Finally, as a major stress sensor, the nucleolus could be the starting point of the death signaling.