The RNA binding protein heterogeneous nuclear ribonucleoprotein L-Like (hnRNP LL) is known to regulate the alternative splicing of various physiologically important precursor messenger RNAs (pre-mRNAs). It undergoes a wide range of post-translational modifications (PTMs), including phosphorylation, ubiquitination and acetylation. However, the target amino acids and effects of these PTMs on the functions of hnRNP LL have not been characterized so far. In this study, we show for the first time that the endogenous hnRNP LL is phosphorylated upon depolarization. Using phosphopeptide mapping followed by the generation of a custom-made phospho-site specific antibody, we further show that phosphorylation at Ser308 of hnRNP LL is induced by depolarization though it is probably not the major phospho-amino acid target of depolarization/CaMKIV. The residue is critical for the nuclear localization and its phosphorylation essential for the CaMKIV-caused perinucleolar localization of the hnRNP LL protein in HEK293T cells. The residue is likely also critical in the regulation of nuclear functions like pre-mRNA splicing. / February 2016

Les protéines hnRNP A1 sont impliquées dans l'épissage alternatif. Un mode d'action proposé implique la formation d'homodimères entre molécules hnRNP A1 causant un réarrangement dans la structure de l'ARN pré-messager. Cette modulation de l'ARN permettrait le rapprochement de sites d'épissage 5' et 3' d'exons situés de par et d'autres d'un exon alternatif. Le domaine riche en résidus glycines est responsable, en grande partie, de l'interaction entre les deux protéines hnRNP A1. Comme la protéine hnRNP H contient aussi un domaine riche en résidus glycines, nous avons postulé que cette dernière pouvait moduler l'épissage alternatif de la même manière que hnRNP A1. Afin de vérifier cette hypothèse, nous avons utilisé un ARN pré-messager constitué de deux sites d'épissage 5' (distal et proximal) en compétition pour un seul site d'épissage 3'. En présence de sites de liaison pour hnRNP H, nous observons que le choix du site d'épissage 5' est déplacé vers le site distal. Nous avons confirmé le rôle des protéines hnRNP H dans la sélection des sites d'épissage 5' in vitro et avons déterminé que le domaine riche en résidus glycines (GRD) est important pour l'activité d'épissage de ce régulateur. Nous avons ensuite exploré la possibilité que des combinaisons de sites de liaison pour hnRNP H et hnRNP A1 puissent activer l'utilisation du site d'épissage 5' distal. Nous avons observé que des combinaisons hétérotypiques peuvent reproduire cette activité d'épissage. Finalement, nous avons utilisé la technologie BRET ("bioluminescence resonance energy transfer") pour démontrer que des interactions homotypiques entre protéines hnRNP H et hétérotypiques entre molécules hnRNP A1 et hnRNP H peuvent se former dans les cellules vivantes. Notre étude suggère que les protéines hnRNP H et hnRNP A1 peuvent changer la conformation de l'ARN pré-messager et affecter le choix du site d'épissage.

<p>The family of proteins known as heterogeneous nuclear ribonucleoproteins (hnRNPs) is large and diverse. Often, one and the same hnRNP will perform multiple cellular functions, leading to their description as “multifunctional proteins”. The two hnRNPs known as hnRNP A1 and hnRNP C1/C2 are multifunctional proteins found to affect the transcription, splicing, stability, and translation of specific genes’ mRNA. They are implicated in carcinogenesis, apoptosis, and DNA damage response mechanisms.</p><p>The aims of this thesis were to study the hnRNP A1 and hnRNP C1/C2 dependent regulation of two highly stress responsive genes, the tumor suppressor p53 and the cytochrome P450 enzyme <i>Cyp2a5/CYP2A6</i>. We identified hnRNP C1/C2 as a DNA damage induced binding protein towards the coding region of p53 mRNA, and found that while a specific <i>cis</i> binding site appears to have a positive function in p53 expression, interaction of hnRNP C1/C2 with this site represses the expression. The data suggest that two distinct molecular mechanisms exist for the down-regulation of p53 by hnRNP C1/C2. One mechanism, active during transcriptional stress, is dependent upon the aforementioned site, and the other, independent. We discuss how hnRNP C1/C2 dependent repression of p53 may play a role in apoptosis. </p><p>The data presented here further suggest that the transcriptional and post-transcriptional processes controlling the expression of the murine <i>Cyp2a5</i> gene are linked <i>via</i> hnRNP A1, by performing functions in the nucleus as a transcription factor, or in the cytoplasmic compartment as a <i>trans </i>factor bound to the 3’UTR of the mRNA as needed. Our studies of the human ortholog of this gene, <i>CYP2A6</i>, suggest that this gene is regulated post-transcriptionally in a manner similar to that of its murine counterpart, <i>via</i> changes in mRNA stability and interaction of hnRNP A1 with its 3’ UTR. </p>

The family of proteins known as heterogeneous nuclear ribonucleoproteins (hnRNPs) is large and diverse. Often, one and the same hnRNP will perform multiple cellular functions, leading to their description as “multifunctional proteins”. The two hnRNPs known as hnRNP A1 and hnRNP C1/C2 are multifunctional proteins found to affect the transcription, splicing, stability, and translation of specific genes’ mRNA. They are implicated in carcinogenesis, apoptosis, and DNA damage response mechanisms. The aims of this thesis were to study the hnRNP A1 and hnRNP C1/C2 dependent regulation of two highly stress responsive genes, the tumor suppressor p53 and the cytochrome P450 enzyme Cyp2a5/CYP2A6. We identified hnRNP C1/C2 as a DNA damage induced binding protein towards the coding region of p53 mRNA, and found that while a specific cis binding site appears to have a positive function in p53 expression, interaction of hnRNP C1/C2 with this site represses the expression. The data suggest that two distinct molecular mechanisms exist for the down-regulation of p53 by hnRNP C1/C2. One mechanism, active during transcriptional stress, is dependent upon the aforementioned site, and the other, independent. We discuss how hnRNP C1/C2 dependent repression of p53 may play a role in apoptosis. The data presented here further suggest that the transcriptional and post-transcriptional processes controlling the expression of the murine Cyp2a5 gene are linked via hnRNP A1, by performing functions in the nucleus as a transcription factor, or in the cytoplasmic compartment as a trans factor bound to the 3’UTR of the mRNA as needed. Our studies of the human ortholog of this gene, CYP2A6, suggest that this gene is regulated post-transcriptionally in a manner similar to that of its murine counterpart, via changes in mRNA stability and interaction of hnRNP A1 with its 3’ UTR.

RNA-binding proteins play a central role in the post-transcriptional regulation of gene expression; however, little is known about the endogenous transcripts to which they bind. Here, I have used the ultra-violet cross-linking and immuno-precipitation (CLIP) technique to identify RNA targets directly bound to two RNA-binding proteins: Acinus and hnRNP A1. Acinus (apoptotic chromatin condensation inducer in the nucleus) contains a region that is homologous to the RNA binding domain of the Drosophila splicing regulator sex-lethal, and a serine and arginine rich region similar to that seen in the SR family of proteins, which function extensively in splicing. Furthermore it is a component of the multi-protein spliceosome complex, and I have demonstrated it can directly bind polyadenylated RNA. I have shown that Acinus displays a diffuse nuclear localisation pattern, however, overexpression of an epitope-tagged protein results in its accumulation in enlarged nuclear speckles. Together these results suggest a role in pre-mRNA splicing. Acinus is cleaved during apoptosis by caspase-3, resulting in a truncated protein with chromatin condensation inducing activity (Sahara et al., 1999). Accordingly, I have demonstrated that overexpression of epitope-tagged Acinus results in an increased number of cells exhibiting an apoptotic phenotype. The proteolytic fragment contains the RNA binding region, and to determine if the role of Acinus in apoptosis is mediated by RNA interactions I utilised CLIP to identify in vivo RNA targets. I have identified several mRNA targets of Acinus and found that the binding sites in those mRNA targets predominantly map to constitutively expressed exons. This is in agreement with the exon junction complex, of which Acinus is a component, being deposited on mRNAs after splicing. These results may indicate that Acinus is a core RNA binding factor of the exon junction complex. To complement this approach, I also performed CLIP with a known alternative splicing regulator, hnRNP A1. In this manner, the binding site preferences could be compared between the two proteins. As expected, the majority of hnRNP A1 binding sites are located in introns, corresponding with their identified role of antagonizing pre-mRNA splicing by binding intronic splicing elements. Interestingly, a number of the CLIP tags are located in, or adjacent to, alternatively spliced events suggesting a role for hnRNP A1 in the regulation of alternative splicing of these specific pre-mRNAs. In addition to pre-mRNA splicing hnRNP A1 also functions in the cellular stress response. Upon environmental stresses it relocates to the cytoplasm and accumulates in cytoplasmic foci known as stress granules (Guil et al., 2006). Here I show some of the targets identified by CLIP are regulated by hnRNP A1 in times of cellular stress. In summary, I have identified two novel subsets of RNAs, bound by Acinus or hnRNP A1 in vivo. I have shown these proteins exhibit distinct binding preferences, which correspond to their biological function. This work is consistent with hnRNP A1 acting as an alternative splicing regulator, and provides evidence for a dual role of Acinus in mRNA splicing and apoptosis. This study also demonstrates the power of the CLIP technique, as identification of in vivo RNA targets allows greater understanding of the mechanisms by which RNA-binding proteins exert their regulatory control.

Trypanosoma cruzi é o agente causador da Doença de Chagas, que segundo a OMS, atinge oito milhões de pessoas principalmente na América Latina, causando danos à saúde pública, juntamente com um impacto econômico negativo. Durante o processo de infecção, uma variedade de eventos celulares ocorre apenas pelo simples contato do parasito com a célula hospedeira, levando a modificações no metabolismo celular e alterações morfológicas. O parasita é capaz de modular respostas celulares e imunológicas da célula hospedeira para sua própria sobrevivência. Além do que, pode alterar compartimentos celulares como o número e tamanho de nucléolos, sugerindo que a presença do parasita poderia estar interferindo na maquinaria nuclear. Porém, pouco se conhece sobre a organização nuclear da célula hospedeira quando infectada por Trypanosoma cruzi. O objetivo deste estudo foi de investigar pela primeira vez os compartimentos nucleares das células hospedeiras durante o curso da infecção por T. cruzi. Células LLC-MK2 foram infectadas com T. cruzi e reações de imunofluorescência indireta foram realizadas utilizando anticorpos e marcadores específicos para proteínas nucleares. As análises das imagens de microscopia confocal e quantificação das fluorescências pelo ImageJ mostraram padrões distintos nos compartimentos nucleares quando comparadas com células não infectadas. Corpos de Cajal e Speckles sofrem alterações quando a célula está infectada e isso depende do ciclo celular do parasita. Neste trabalho também foi investigado através de quantificação de imagem e immunoblotting o comportamento das Ribonucleoproteínas A1 e A2B1 durante a parasitemia. Estas análises demonstram que o T. cruzi pode modular a célula hospedeira quando infectada a favor de sua sobrevivência, promovendo alterações na dinâmica dos compartimentos nucleares durante o seu ciclo celular. Esse estudo inédito poderá auxiliar a compreender a biologia do parasita e sua interação com a célula hospedeira e desta maneira contribuir na busca de possíveis alvos terapêuticos / Trypanosoma cruzi is the causal agente of Chagas disease, that affects about eight million people mostly in Latin America according to the WHO, causing damage to public health and a negative economic impact. During infection, a variety of signaling processes occur after contact of the parasite to the host cell, what can lead to metabolic modifications as well morphological alterations in both cells. The parasite can modulate host cell cellular and immunological responses for its own survival. In addiction, the presence of T. cruzi can modify the nuclear compartments such as nucleoli, suggesting that the presence of the parasite could be interfering with the nuclear machinery. However, little is know about the nuclear organization when the host cell is infected with Trypanosoma cruzi. This study aimed to investigate for the first time the nuclear compartment of host cells infected by T. cruzi using specific antibodies and fluorescent markers for nuclear compartments, in order to investigate the morphological and functional changes in the nucleus of the host cell. Using LLC-MK2 cells infected with T. cruzi, we performed indirect immunofluorescence using distinct nuclear antibodies. Confocal microscopy analysis of infected cells showed pattern variations in the nuclear compartments when compared to uninfected cells. Cajal bodies and Speckles suffer alterations when the cell is infected and it is related to the parasite life cycle. In this work we also investigated by image quantification and immunoblotting the behavior of Ribonucleoproteins A1 and A2B1 during infection. These evidences support the idea that T. cruzi can modulate host cell response to ensure its own survival during the infection, promoting changes in the dynamics of the nuclear compartments. This unpublished data may help to understand the biology of the parasite and its interaction with the host cell and thus contributing to seek for potential therapeutic targets

L'épissage est une étape clef de la multiplication du VIH-1. Par utilisation de 4 sites donneurs et 8 sites accepteurs d'épissage, plus de 40 ARNm différents sont produits. Une approche protéomique nous a permis d'identifier de nouvelles protéines interagissant avec la région de l'ARN viral contenant le site A7. Nous avons démontré l'interaction directe avec l'ARN viral de 5 des protéines identifiées (nucléoline, hnRNP A1/B, hnRNP H et hnRNP K). Nous avons montré que hnRNP K a plusieurs sites de fixation dans la région du site A7 et que hnRNP A1et hnRNP K se lient de façon coopérative. Nous avons montré un effet inhibiteur de hnRNP K sur l'épissage au site A7. Comme la protéine hnRNP A1 est un régulateur négatif de plusieurs sites accepteurs d'épissage (A1, A2, A3, A7), nous avons testé si la protéine hnRNP K pouvait renforcer l'inhibition à ces sites. En fait, hnRNP K active l'épissage in vitro des introns entre le site donneur D1 et les sites accepteurs A1, A2 et A3. Nous avons montré que la protéine hnRNP K renforce fortement l'activité de ASF/SF2 au site A2, ce qui indique que selon le contexte, la protéine hnRNP K peut être activatrice ou inhibitrice de l'épissage du VIH-1. J'ai observé de plus que la surexpression de la protéine hnRNP K dans des cellules HeLa, transfectées avec le plasmide p PSP contenant le virus VIH-1 dépourvu de ses capacités d'encapsidation, produit un changement très marqué de l'épissage alternatif de l'ARN PSP, ce qui confirme la forte influence de hnRNP K sur l'épissage alternatif du VIH-1. L'augmentation de la concentration cellulaire de hnRNP K dans les cellules HeLa conduit aussi à une diminution de la protéine virale Nef. La protéine hnRNP K intervient donc non seulement dans la régulation du site A7, mais aussi dans celle de la majorité des sites d'épissage régulés de l'ARN du VIH. L'action de cette protéine sur plusieurs des sites d'épissage montre que la protéine hnRNP K est probablement un régulateur général de l'épissage de VIH-1 / HIV-1 pre-mRNA splicing depends upon 4 donor and 8 acceptor sites, which are used in combination to produce more than 40 different mRNAs. To further characterize nuclear factors involved in these processes, we purified RNP complexes formed by incubation of SLS2-A7 transcripts in HeLa cell nuclear extracts by affinity chromatography to identify new associated proteins. We showed that, in addition to the well known hnRNP A1 inhibitor of site A7, nucleolin, hnRNP H and hnRNP K interact directly with SLS2-A7 RNA. We demonstrated that hnRNP K has multiple binding sites in the vicinity of site A7 and that binds cooperatively to hnRNP A1 to the A7 RNA region and limits the A7 utilization in vitro. As hnRNP A1 is a negative regulator of several HIV-1 splicing sites (A1, A2, A3), we tested whether hnRNP K may also reinforce hnRNP A1 inhibition at these sites. Surprisingly, hnRNP K activated in vitro splicing of the D1-A1, D1-A2 and D1-A3 introns. Interestingly, hnRNP K was found to reinforce strongly the ASF/SF2 activity at site A2, which indicates that depending on the splicing site hnRNP K can be a splicing activator or inhibitor. To test how hnRNP K influences the relative utilization of HIV-1 splicing sites in cellulo, we used plasmid p PSP containing all the HIV-1 splicing sites and tested the effect of over-expression in HeLa cells on alternative splicing of the PSP RNA. Doubling the amount of hnRNP K in HeLa cells led to a drastic change of the PSP RNA alternative splicing, which confirms the strong influence of hnRNP K on alternative splicing. Moreover, increase of cellular concentration of hnRNP K strongly decrease the viral Nef protein production. hnRNP K protein affects A7 splicing regulation but also regulates the majority of regulated splicing sites of HIV. By extension of the study of hnRNP K effect to other HIV-1 splicing sites, we discovered that hnRNP K is a general regulator of HIV-1 splicing

Human Rhinoviruses (HRVs) utilize Internal Ribosome Entry Sites (IRES) to drive viral protein synthesis. IRESs are specialized RNA elements present within the 5’ UTR of mRNAs that recruit ribosomes independently of the 5’ m7G cap structure. hnRNP A1 (heterogeneous nuclear ribonucleoprotein A1), a multifunctional RNA binding protein, is required for the IRES-dependent translation of many specific RNAs within the cell cytoplasm. The phosphorylation of hnRNP A1 is required for its cytoplasmic accumulation. I have identified and validated the role of HK2 in hnRNP A1 cellular localization by immunofluorescence microscopy, by analysis of HRV infection and by siRNA-based screening. These studies show that decreased HK2 protein levels lead to decreased cytoplasmic accumulation of hnRNPA1 during osmotic shock and HRV infection, to a decrease in HRV-infected cells and to decreased caspase activation in osmotically stressed and HRV-infected cells. Thus, HK2 may regulate hnRNP A1 cytoplasmic localization following HRV infection.

L'épissage alternatif des pré-ARN messagers est un mécanisme qui permet de générer une très grande diversité protéique chez les eucaryotes supérieurs. La sélection des sites d'épissage permet ainsi de produire certains isoformes protéiques plutôt que d'autres dans des conditions précises. Cette modulation implique généralement la participation d'une multitude de facteurs aux propriétés parfois synergiques et/ou antagonistes. Dans le cas du pré-ARN messager hnRNP A1, au moins trois éléments distincts renforcent l'exclusion de l'exon 7B. Par contre, l'élément intronique conservé de 38 nt (CE9) situé en aval de l'exon 7B permet la répression du site d'épissage 3', ce qui entraînerait l'inclusion de l'exon 7B. La portion 5' de l'élément CE9 est liée par la protéine SRp30c et cette interaction est importante pour permettre l'activité de CE9 in vitro. Afin de déterminer les composantes essentielles à l'activité de répression de SRp30c, des sites de liaison de haute affinité pour cette protéine ont été identifiés à l'aide de la procédure"SELEX". Les résultats obtenus indiquent que plusieurs sites de haute affinité reproduisent l'activité de l'élément CE9 complet dans un essai d'épissage où deux sites d'épissage 3' sont en compétition pour un seul site d'épissage 5'. De plus, les résultats suggèrent une contribution de la portion 3' de CE9, en plus de la partie 5', pour la liaison de la protéine SRp30c. En utilisant la séquence complète de CE9 pour effectuer une chromatographie d'affinité, la protéine hnRNP UPTB a été isolée et identifiée par spectrométrie de masse. Cependant, nous avons été surpris de constater que la protéine recombinante PTB agit comme activateur du site d'épissage 3' en diminuant l'activité de répression de CE9. Ces résultats suggèrent donc un nouveau rôle pour la protéine PTB, c'est-à-dire comme anti-répresseure de l'activité d'inhibition de SRp30c. PTB pourrait donc être un nouveau facteur capable de contrôler l'épissage alternatif du pré-ARN messager de hnfP A1.

Thèse numérisée par la Direction des bibliothèques de l'Université de Montréal.