The ability of proteins to engage in nuclear-cytoplasmic shuttling is required for their proper function. The nuclear export of the von Hippel Lindau (VHL) tumour suppressor protein is necessary for the proteasomal degradation of the hypoxia inducible factor alpha (HIFα). Studies have identified that the nuclear export of VHL and other proteins encoding a Transcription-Dependent Nuclear Export Motif (TD-NEM) is independent of the classical CRM1 nuclear export pathway but requires ongoing transcription. Furthermore, the eukaryotic elongation factor 1 alpha (eEF1A) was identified as a mandatory component of the TD-NEM-mediated nuclear export machinery. In this study, we have uncovered the ability of eEF1A to mediate the nuclear export of proteins by accessing the nuclear compartment in its inactive, GDP-bound form. Although previously thought of as a strictly cytoplasmic protein, work conducted in this thesis has shown that eEF1A is a nuclear-cytoplasmic shuttling protein and this ability is required for the effective export of proteins encoding a TD-NEM.

The ability of proteins to engage in nuclear-cytoplasmic shuttling is required for their proper function. The nuclear export of the von Hippel Lindau (VHL) tumour suppressor protein is necessary for the proteasomal degradation of the hypoxia inducible factor alpha (HIFα). Studies have identified that the nuclear export of VHL and other proteins encoding a Transcription-Dependent Nuclear Export Motif (TD-NEM) is independent of the classical CRM1 nuclear export pathway but requires ongoing transcription. Furthermore, the eukaryotic elongation factor 1 alpha (eEF1A) was identified as a mandatory component of the TD-NEM-mediated nuclear export machinery. In this study, we have uncovered the ability of eEF1A to mediate the nuclear export of proteins by accessing the nuclear compartment in its inactive, GDP-bound form. Although previously thought of as a strictly cytoplasmic protein, work conducted in this thesis has shown that eEF1A is a nuclear-cytoplasmic shuttling protein and this ability is required for the effective export of proteins encoding a TD-NEM.

The ability of proteins to engage in nuclear-cytoplasmic shuttling is required for their proper function. The nuclear export of the von Hippel Lindau (VHL) tumour suppressor protein is necessary for the proteasomal degradation of the hypoxia inducible factor alpha (HIFα). Studies have identified that the nuclear export of VHL and other proteins encoding a Transcription-Dependent Nuclear Export Motif (TD-NEM) is independent of the classical CRM1 nuclear export pathway but requires ongoing transcription. Furthermore, the eukaryotic elongation factor 1 alpha (eEF1A) was identified as a mandatory component of the TD-NEM-mediated nuclear export machinery. In this study, we have uncovered the ability of eEF1A to mediate the nuclear export of proteins by accessing the nuclear compartment in its inactive, GDP-bound form. Although previously thought of as a strictly cytoplasmic protein, work conducted in this thesis has shown that eEF1A is a nuclear-cytoplasmic shuttling protein and this ability is required for the effective export of proteins encoding a TD-NEM.

Eukaryotic Elongation Factor 1A (eEF1A) exists in two forms in vertebrates. The first form, eEF1A1, is expressed ubiquitously throughout development but is downregulated postdevelopmentally and replaced with eEF1A2, an isoform sharing 92% amino acid identity, in neurons and muscle. The primary function of eEF1A is to recruit amino-acylated tRNAs in a GTP-dependent manner to the A site of the ribosome during protein translation, but it also has non-canonical roles in the cell, some of which are isoform dependent. The reasons for the cell-type dependent switch from eEF1A1 to eEF1A2 are poorly understood. The first aim of this project was to examine the role played by eEF1A isoforms in neuritogenesis. To do this I used RNAi to significantly reduce expression of one or other isoform in neuronal cells and measure the effects this had on neurite outgrowth. Neurite outgrowth was significantly reduced in cells depleted of eEF1A1, but not eEF1A2. The complete loss of eEF1A2 is fatal, as has been demonstrated in the wasted mouse, an eEF1A2-null model characterised by muscle wastage, neurodegeneration and death at 4 weeks of age. Mice heterozygous for the wasted mutation have normal motor function. Recent work has found that heterozygous missense mutations in eEF1A2 can cause epilepsy and intellectual disability. It is not yet known whether the seven different de novo mutations identified to date confer loss or gain of function – a crucial piece of information required before possible treatments can be sought. The second aim of this project therefore was to investigate the role of eEF1A2 in epilepsy and intellectual disability. I achieved this by using CRISPR in two ways; firstly to model one of the mutations, D252H, in vitro in a neuronal cell line, and secondly to model another of the mutations, G70S, in vivo. No mice that recapitulated the human disease condition of EEF1A2G70S/+ were obtained however, due to the error-prone nature of the non-homologous end joining repair pathway activated by CRISPR-mediated DNA cleavage, 17 of the 35 mice born were found to be homozygous nulls at the Eef1a2 locus. Nine of these had fatal audiogenic seizures caused by sudden loud noises within the animal unit. Three mice were Eef1a2G70S/- and one Eef1a2G70S/G70S but these nonetheless showed a wasted phenotype, indicating that this mutant form of eEF1A2 has compromised function, at least in terms of translation elongation. Whether it has a toxic function ca not yet be known, but the severity of the phenotype in the G70S homozygous animal could suggest a gain of function. In in vitro experiments with exogenous eEF1A2 carrying the epilepsy-causing mutation R423C, protein expression of the mutant construct in immortalised cell lines was significantly higher when cotransfected with the wildtype construct, which mirrors the condition in humans, than when transfected alone, so the mutant protein could be stabilised in the presence of wildtype eEF1A2. I used CRISPR on LUHMES cells to make a mutant neuronal cell line containing the D252H mutation in eEF1A2. Due to time restraints no phenotypic differences between the wild type line and the D252H mutation line have yet been identified, but would form the focus of a future project.

No description available.

Abnormal expression of the eukaryotic translation elongation factor 1A (eEF1A) has been implicated in disease states such as motor neuron degeneration and cancer. Two variants of eEF1A are found in mammals, named eEF1A1 and eEF1A2. These two variants are encoded by different genes, produce proteins which are 92% identical but have very different patterns of expression. eEF1A1 is almost ubiquitously expressed while eEF1A2 is expressed only in specialised cell types such as motor neurons and muscle. A spontaneous mutation in eEF1A2 results in the wasted mouse phenotype which shows similar characteristics in the mouse to those seen in human motor neuron degeneration. This mutation has been shown to be a 15.8kb deletion resulting in the complete loss of the promoter region and first non coding exon of eEF1A2 which completely abolishes protein expression. The main aim of this project was to further investigate the role of eEF1A2 in motor neuron degeneration. Firstly, although the wasted phenotype is considered to be caused by a recessive mutation, I established a cohort of aged heterozygote mice to evaluate whether any changes are seen later in life that might model late onset motor neuron degeneration. A combination of behavioural tests and pathology was used to compare wild type and heterozygous mice up to 21 months of age. Whilst results indicate that there is no significant difference between ageing heterozygotes and wildtype controls, there is an indication that female heterozygote mice perform slightly worse that wildtype controls on the rotarod (a behavioural test for motor function). Secondly, I aimed to investigate the primary cause of the wasted pathology by generating transgenic wasted mice expressing neuronal eEF1A2 only. This would complement previous experiments in the lab which studied transgenic wasted mice expressing eEF1A2 in muscle only. Unfortunately the expression of eEF1A2 in the transgenic animals was not neuronal specific. However a transgenic line with expression of eEF1A2 in neurons and skeletal muscle but not cardiac muscle has been generated which clearly warrants further investigation. Thirdly, I wished to assess whether eEF1A2 has any role in human motor neuron degeneration. To achieve this, eEF1A2 expression was investigated in spinal cords from human motor neuron disease (MND) patients. Preliminary data suggests that motor neurons from some MND patients express significantly less eEF1A2 than motor neurons of control samples. Further work is required to confirm these findings. Finally, I investigated the individual roles of eEF1A1 and eEF1A2 in the heat shock response. I used RNAi to ablate each variant separately in cells and subsequently measured the ability of each variant individually to mount a heat shock response. Results indicate a clear role for eEF1A1 but not eEF1A2 in the induction of heat shock. This may explain in part why motor neurons exhibit a poor heat shock response as they express eEF1A2 and not eEF1A1. These experiments shed light on our understanding of the role of eEF1A2 in motor neuron degeneration and uncover many new avenues of future investigation.

Eukaryotic Elongation Factor 1A (eEF1A) plays a key role in protein synthesis by delivering aminoacylated tRNAs to the A site of the ribosome. In higher vertebrates, two isoforms of eEF1A exist called eEF1A1 and eEF1A2, with eEF1A2 being expressed in adult brain, heart and skeletal muscle. Since 2012, several different de novo heterozygous missense mutations in EEF1A2 have been identified in humans and these cause epilepsy, intellectual disability and autism. Before considering treatment options, it is vital to determine whether these mutations cause loss or gain of protein function. I performed a battery of behavioural tests using two mouse lines with heterozygous loss of function mutations in eEF1A2. The aim was to determine whether there were any behavioural phenotypes consistent with intellectual disability and/or autism. Using heterozygous wasted mice (Eef1a2+/wst), I analysed the effects of aging on behaviour and found that Eef1a2+/wst mice showed reduced marble burying activity and reduced movement in the open field test with age. In a test of social behaviour, Eef1a2+/wst mice showed a significantly reduced preference for social novelty at all ages tested. The second heterozygous null line, Del22.ex3, was generated on a pure C57BL/6J genetic background. This new line was made in order to reduce the level of variation observed in data from the wasted line, which was on a mixed genetic background. The genetic background was shown to have an influence on behaviour as the results differed between this line and the wasted line. Del22.ex3 Eef1a2+/- mice showed significantly reduced engagement in repetitive behaviours compared with wild-type littermates and normal preference for social novelty. Using CRISPR/Cas9, a mouse line with the D252H missense mutation was generated and I repeated my behavioural testing on heterozygotes from this line. I found no behavioural abnormalities in this line suggesting a mouse-human difference in the ability to tolerate eEF1A2 missense mutations. Previous attempts to make a line with the G70S missense mutation were unsuccessful but as a product of this experiment, it was found that mice expressing G70S eEF1A2 had a comparable phenotype to and died at the same age as complete knockouts. This suggested that the G70S protein is non-functional and cannot compensate for loss of wild-type eEF1A2. These experiments have improved our understanding of the phenotypic effects of Eef1a2 mutations in mice and have shown, for the first time, that mutations in Eef1a2 affect mouse behaviour.

L’auto-incompatibilité (AI) est une barrière reproductive prézygotique qui permet aux pistils d’une fleur de rejeter leur propre pollen. Les systèmes d’AI peuvent prévenir l’autofertilisation et ainsi limiter l’inbreeding. Dans l’AI gamétophytique, le génotype du pollen détermine son propre phénotype d’incompatibilité, et dans ce système, les déterminants mâles et femelles de l’AI sont codés par un locus multigénique et multi-allélique désigné le locus S. Chez les Solanaceae, le déterminant femelle de l’AI est une glycoprotéine stylaire extracellulaire fortement polymorphique possédant une activité ribonucléase et désignée S-RNase. Les S-RNases montrent un patron caractéristique de deux régions hypervariables (HVa et HVb), responsables de leur détermination allélique, et cinq régions hautement conservées (C1 à C5) impliquées dans l’activité catalytique ou la stabilisation structurelle de ces protéines. Dans ce travail, nous avons investigué plusieurs caractéristiques des S-RNases et identifié un nouveau ligand potentiel aux S-RNases chez Solanum chacoense. L’objectif de notre première étude était l’élucidation du rôle de la région C4 des S-RNases. Afin de tester l’hypothèse selon laquelle la région C4 serait impliquée dans le repliement ou la stabilité des S-RNases, nous avons généré un mutant dans lequel les quatre résidus chargés présents en région C4 furent remplacés par des résidus glycine. Cette protéine mutante ne s’accumulant pas à des niveaux détectables, la région C4 semble bien avoir un rôle structurel. Afin de vérifier si C4 est impliquée dans une liaison avec une autre protéine, nous avons généré le mutant R115G, dans lequel un acide aminé chargé fût éliminé afin de réduire les affinités de liaison dans cette région. Ce mutant n’affectant pas le phénotype de rejet pollinique, il est peu probable que la région C4 soit impliquée dans la liaison des S-RNases avec un ligand ou leur pénétration à l’intérieur des tubes polliniques. Enfin, le mutant K113R, dans lequel le seul résidu lysine conservé parmi toutes les S-RNases fût remplacé par un résidu arginine, fût généré afin de vérifier si cette lysine était un site potentiel d’ubiquitination des S-RNases. Toutefois, la dégradation des S-RNases ne fût pas inhibée. Ces résultats indiquent que C4 joue probablement un rôle structurel de stabilisation des S-RNases. Dans une seconde étude, nous avons analysé le rôle de la glycosylation des S-RNases, dont un site, en région C2, est conservé parmi toutes les S-RNases. Afin d’évaluer la possibilité que les sucres conjugués constituent une cible potentielle d’ubiquitination, nous avons généré une S11-RNase dont l‘unique site de glycosylation en C2 fût éliminé. Ce mutant se comporte de manière semblable à une S11-RNase de type sauvage, démontrant que l’absence de glycosylation ne confère pas un phénotype de rejet constitutif du pollen. Afin de déterminer si l’introduction d’un sucre dans la région HVa de la S11-RNase pourrait affecter le rejet pollinique, nous avons généré un second mutant comportant un site additionnel de glycosylation dans la région HVa et une troisième construction qui comporte elle aussi ce nouveau site mais dont le site en région C2 fût éliminé. Le mutant comportant deux sites de glycosylation se comporte de manière semblable à une S11-RNase de type sauvage mais, de manière surprenante, le mutant uniquement glycosylé en région HVa peut aussi rejeter le pollen d’haplotype S13. Nous proposons que la forme non glycosylée de ce mutant constitue un allèle à double spécificité, semblable à un autre allèle à double spécificité préalablement décrit. Il est intéressant de noter que puisque ce phénotype n’est pas observé dans le mutant comportant deux sites de glycosylation, cela suggère que les S-RNases ne sont pas déglycosylées à l’intérieur du pollen. Dans la dernière étude, nous avons réalisé plusieurs expériences d’interactions protéine-protéine afin d’identifier de potentiels interactants polliniques avec les S-RNases. Nous avons démontré que eEF1A, un composant de la machinerie de traduction chez les eucaryotes, peut lier une S11-RNase immobilisée sur résine concanavaline A. Des analyses de type pull-down utilisant la protéine eEF1A de S. chacoense étiquetée avec GST confirment cette interaction. Nous avons aussi montré que la liaison, préalablement constatée, entre eEF1A et l’actine est stimulée en présence de la S11-RNase, bien que cette dernière ne puisse directement lier l’actine. Enfin, nous avons constaté que dans les tubes polliniques incompatibles, l’actine adopte une structure agrégée qui co-localise avec les S-RNases. Ces résultats suggèrent que la liaison entre eEF1A et les S-RNases pourrait constituer un potentiel lien fonctionnel entre les S-RNases et l’altération du cytosquelette d’actine observée lors des réactions d’AI. Par ailleurs, si cette liaison est en mesure de titrer les S-RNases disponibles à l’intérieur du tube pollinique, ce mécanisme pourrait expliquer pourquoi des quantités minimales ou « seuils » de S-RNases sont nécessaires au déclenchement des réactions d’AI. / Self-incompatibility (SI) is a prezygotic reproductive barrier that allows the pistil of a flower to specifically reject their own (self-) pollen. SI systems can help prevent self-fertilization and avoid inbreeding. In gametophytic SI (GSI), the genotype of the pollen determines its breeding behaviour and in this system both female and male specificity determinants of SI are under the control of a multigenic and multiallelic locus called the S-locus. In Solanaceae, the female determinant of SI is a highly polymorphic stylar-expressed extracellular glycoprotein with RNase activity called the S-RNase. S-RNases show a distinct pattern of two hypervariable (HVa and HVb) regions, responsible for their allelic specificity, and five highly conserved regions (C1 to C5) thought to be involved in either the catalytic activity or the structural stabilization of the protein. In this work, we analyzed and characterized several conserved features of the S-RNases and also identified a potential novel S-RNase interactant in Solanum chacoense. The aim of our first study was to investigate the role of the C4 region of S-RNases. To test the hypothesis that the C4 region may be involved in S-RNase folding or stability, we examined a mutant in which the four charged residues in the C4 region were replaced with glycine. This mutant did not accumulate to detectable levels in styles, supporting a structural role for C4. To test the possibility that C4 might be involved in binding another protein, we prepared an R115G mutant, in which a charged amino acid was eliminated to reduce any potential binding to this region. This mutant had no effect on the pollen rejection phenotype of the protein, and thus C4 is likely not involved in either ligand binding or S-RNase entry inside pollen tubes. Finally, a K113R mutant, in which the only conserved lysine residue in all the S-RNases was replaced with arginine, was generated to test if this residue was an S-RNase ubiquitination site. However, S-RNase degradation was not disrupted in this mutant. Taken together, these results indicate that the C4 region likely plays a structural role. In a second study, we analyzed the role of S-RNase glycosylation. All S-RNases share a conserved glycosylation site in the C2 region. To test the possibility that the sugar residues might be a target for ubiquitination, a transgenic S11-RNase lacking its single glycosylation site was examined. This construct behaved similarly to a wild type S11-RNase, demonstrating that the lack of glycosylation does not confer constitutive pollen rejection. To determine if the introduction of an N-linked glycan in the HVa region would affect pollen rejection, a construct containing a second N-glycosylation site inside the HVa region of the S11-RNase and a construct containing only that N-glycosylation site inside the HVa region were prepared. The first construct rejected S11 pollen normally, but surprisingly, plants expressing the construct lacking the C2 glycosylation site rejected both S11 and S13 pollen. We propose that the non-glycosylated form is a dual specific allele, similar to a previously described dual-specific allele that also had amino acid replacements in the HV regions. Interestingly, this phenotype is not observed in the mutant containing two glycosylation sites, which suggests that the sugar residues are not removed during S-RNase entry into the pollen. In the final study, S-RNase-binding assays were performed with pollen extracts to detect potential interacting proteins. We found that concanavalin A-immobilized S11-RNase bound eEF1A, a component of the eukaryotic translational machinery. This interaction was validated by pull-down experiments using a GST-tagged S. chacoense eEF1A. We also found that a previously documented actin binding to eEF1A was markedly increased in the presence of S-RNases, although S-RNases alone do not bind actin. Lastly, we observed that actin in incompatible pollen tubes has an unusual aggregated form which also co-labels with S-RNases. This suggests that binding between S-RNases and eEF1A could provide a potential functional link between the S-RNase and the alteration of the actin cytoskeleton that occurs during the SI reaction. Furthermore, if eEF1A binding to S-RNases acted to titrate the amount of free S-RNase in the pollen tube, this binding may help explain the threshold phenomenon, where a minimum quantity of S-RNase in the style is required to trigger the SI reaction.