Histone deacetylases (HDAC) 1 and 2 play crucial role in chromatin remodeling and gene expression regimes, as part of multiprotein corepressor complexes. Protein kinase CK2-driven phosphorylation of HDAC1 and 2 regulates their catalytic activities and is required to form the corepressor complexes. Phosphorylation-mediated differential distributions of HDAC1 and 2 complexes in regulatory and coding regions of transcribed genes catalyze the dynamic protein acetylation of histones and other proteins, thereby influence gene expression. During mitosis, highly phosphorylated HDAC1 and 2 heterodimers dissociate and displace from mitotic chromosomes. Our goal was to identify the kinase involved in mitotic phosphorylation of HDAC1 and 2. We postulated that CK2-mediated increased phosphorylation of HDAC1 and 2 leads to dissociation of the heterodimers, and, the mitotic chromosomal exclusions of HDAC1 and 2 are largely due to the displacement of HDAC-associated proteins and transcription factors, which recruit HDACs, from chromosomes during mitosis. We further explored the role of un- or monomodified HDAC1 and 2 complexes in immediate-early genes (IEGs), FOSL1 (FOS-like antigen-1) and MCL1 (Myeloid cell leukemia-1), regulation. Dynamic histone acetylation is an important regulator of these genes that are overexpressed in a number of diseases and cancers. We hypothesized that transcription dependent recruitment of HDAC1 and 2 complexes over the gene body regions plays a regulatory role in transcription and splicing regulation of these genes. We present evidence that CK2-catalyzed increased phosphorylation of HDAC1 and 2 regulates the formation of distinct corepressor complexes containing either HDAC1 or HDAC2 homodimers during mitosis, which might target cellular factors. Furthermore, the exclusion of HDAC-recruiting proteins is the major factor for their displacement from mitotic chromosomes. We further demonstrated that un- or monophosphorylated HDAC1 and 2 are associated with gene body of FOSL1 in a transcription dependent manner. However, HDAC inhibitors prevented FOSL1 activation independently of the nucleosome response pathway, which is required for IEG induction. Interestingly, our mass spectrometry results revealed that HDAC1 and 2 interact with a number of splicing proteins, in particular, with serine/arginine-rich splicing factor 1 (SRSF1). HDAC1 and 2 are co-occupied with SRSF1 over gene body regions of FOSL1 and MCL1, regardless of underlying splicing mechanisms. Using siRNA-mediated knockdown approaches and HDAC inhibitors, we demonstrated that alternative splicing of MCL1 is regulated by RNA-directed localized changes in the histone acetylation levels at the alternative exon. The change in histone acetylation levels correlates with the increased transcription elongation and results in change in MCL1 splicing by exon skipping mechanism. Taken together, our results contribute to further understanding of how the multi-faceted HDAC1 and 2 complexes can be regulated and function in various processes, including, but not limited to, transcription regulation and alternative splicing. This can be an exciting area of future research for therapeutic interventions.

The RAS/RAF/MEK/ERK pathway is the prototypical cellular signal transduction cascade and has been the focus of intense scrutiny over the last two decades. As a mitogenic pathway, its activation is a potent driver of cellular growth and survival, and its deregulation underlies many cancers. While RAS family GTPases have long been recognized as prolific human oncogenes, a landmark study in 2002 also established the RAF family kinase as a bona fide oncogene (Davies et al., 2002). Indeed, aberrant RAS-RAF signaling underlies nearly one-third of all human cancers (Wellbrock et al., 2004). Notably, mutations in RAF are found with astounding frequency in certain cancers (e.g. 70% of malignant melanomas) (Dhomen and Marais, 2007). These findings have identified intercepting aberrant RAF function as an ideal therapeutic target. RAF is a Ser/Thr protein kinase and its activity is strictly regulated by a core complex of at least three proteins, namely, KSR, CNK and HYP (Claperon and Therrien, 2007). The mechanism by which the KSR/CNK/HYP complex regulates RAF function remains enigmatic. In particular, the function of KSR in regulating RAF activity is highly controversial. The work described in this thesis was conducted with the aim of: i) understanding the interactions that underlie formation of the KSR/CNK/HYP complex, and ii) elucidating the mechanism by which the complex regulates RAF function. I have attempted to accomplish these aims using a combination of structural biology, biochemistry and cell biology approaches. I begin by presenting the structure of the SAM domain mediated interaction between CNK and HYP. I describe a model for how the CNK/HYP interaction in turn serves to recruit KSR to form the higher-order KSR/CNK/HYP complex. Subsequently, I describe the allosteric mechanism by which KSR controls RAF activation via the formation of specific side-to-side kinase domain heterodimers of KSR and RAF. Lastly, I describe a potential mechanism by which RAS directly mediates the attainment of the side-to-side dimer configuration of RAF through its own ability to form dimers. The acquisition of the side-to-side dimer configuration is essential for aberrant RAF signaling in cancers, suggesting future RAF inhibition strategies could be aimed at preventing dimer formation.

Protein kinases comprise a fundamental class of cell function regulators that modify proteins by transferring phosphate groups from a nucleoside triphosphate such as ATP to specific amino acid residues on target proteins, altering protein conformation, function, and activity. As such, protein kinases are major regulators of many biological processes, including gene expression, which consists of the transfer of hereditary information in two major processing steps, transcription of DNA into a complementary precursor RNA transcript (pre-mRNA) and its subsequent translated into protein by the ribosome, where it can then go on to perform various processes in the cell. One particular family of protein kinases, otherwise known as serine/arginine protein-specific protein kinases (SRPKs), is conserved throughout eukaryotes and has been shown to be important in regulating gene expression, yet their roles in the gene expression pathway have yet to be elucidated. SRPK are known to phosphorylate serine/arginine (SR) splicing factor proteins, which are involved in mRNA splice site recognition and recruitment of splicing machinery. Members of the LAMMER kinase subfamily of SRPKs have also been shown to be required for efficient pre-mRNA splicing and important for mediating cellular progression through the cell cycle. To determine what other roles SRPKs play in mRNA processing, it is of use to study the homologous SRPK and LAMMER kinases in fission yeast, S. pombe, Dsk1 and Kic1, respectively. S. pombe provides a genetically valuable model for studying kinase function in RNA processing as both RNA processing machinery and SRPKs are conserved through higher eukaryotes. Using a novel green fluorescent protein tagging system based on properties of the MS2 bacteriophage genome, we are able to label specific mRNA transcripts of interest and visualize their locations in the cell using fluorescence microscopy. By visualizing the mRNA trafficking patterns of intron-containing and intronless mRNA transcripts, we show for the first time that deletions of the Dsk1 and Kic1 genes result in the nuclear retention of mRNA, such that Dsk1 and Kic1 are distinctly involved in mRNA export out of the nucleus.

There is no abstract available for this dissertation. / School of Physical Education

Histone deacetylases (HDAC) 1 and 2 play crucial role in chromatin remodeling and gene expression regimes, as part of multiprotein corepressor complexes. Protein kinase CK2-driven phosphorylation of HDAC1 and 2 regulates their catalytic activities and is required to form the corepressor complexes. Phosphorylation-mediated differential distributions of HDAC1 and 2 complexes in regulatory and coding regions of transcribed genes catalyze the dynamic protein acetylation of histones and other proteins, thereby influence gene expression. During mitosis, highly phosphorylated HDAC1 and 2 heterodimers dissociate and displace from mitotic chromosomes. Our goal was to identify the kinase involved in mitotic phosphorylation of HDAC1 and 2. We postulated that CK2-mediated increased phosphorylation of HDAC1 and 2 leads to dissociation of the heterodimers, and, the mitotic chromosomal exclusions of HDAC1 and 2 are largely due to the displacement of HDAC-associated proteins and transcription factors, which recruit HDACs, from chromosomes during mitosis. We further explored the role of un- or monomodified HDAC1 and 2 complexes in immediate-early genes (IEGs), FOSL1 (FOS-like antigen-1) and MCL1 (Myeloid cell leukemia-1), regulation. Dynamic histone acetylation is an important regulator of these genes that are overexpressed in a number of diseases and cancers. We hypothesized that transcription dependent recruitment of HDAC1 and 2 complexes over the gene body regions plays a regulatory role in transcription and splicing regulation of these genes. We present evidence that CK2-catalyzed increased phosphorylation of HDAC1 and 2 regulates the formation of distinct corepressor complexes containing either HDAC1 or HDAC2 homodimers during mitosis, which might target cellular factors. Furthermore, the exclusion of HDAC-recruiting proteins is the major factor for their displacement from mitotic chromosomes. We further demonstrated that un- or monophosphorylated HDAC1 and 2 are associated with gene body of FOSL1 in a transcription dependent manner. However, HDAC inhibitors prevented FOSL1 activation independently of the nucleosome response pathway, which is required for IEG induction. Interestingly, our mass spectrometry results revealed that HDAC1 and 2 interact with a number of splicing proteins, in particular, with serine/arginine-rich splicing factor 1 (SRSF1). HDAC1 and 2 are co-occupied with SRSF1 over gene body regions of FOSL1 and MCL1, regardless of underlying splicing mechanisms. Using siRNA-mediated knockdown approaches and HDAC inhibitors, we demonstrated that alternative splicing of MCL1 is regulated by RNA-directed localized changes in the histone acetylation levels at the alternative exon. The change in histone acetylation levels correlates with the increased transcription elongation and results in change in MCL1 splicing by exon skipping mechanism. Taken together, our results contribute to further understanding of how the multi-faceted HDAC1 and 2 complexes can be regulated and function in various processes, including, but not limited to, transcription regulation and alternative splicing. This can be an exciting area of future research for therapeutic interventions.

The incidence of chronic kidney disease (CKD) is increasing. CKD is characterized by a gradual decrease in renal function leading to end stage renal disease (ESRD). Damage to the glomerular podocytes, is one of the first hallmarks of CKD. We hypothesized that podocyte prostaglandin E2 (PGE2) receptors contribute to the progression of glomerular injury in models of CKD. To test this hypothesis, transgenic mice were generated with either podocyte-specific overexpression or deletion of the PGE2 EP4 receptor (EP4pod+and EP4pod-/- respectively). Mice were next tested in the 5/6 nephrectomy (5/6 Nx) or angiotensin II (Ang II) models of CKD. These studies revealed increased proteinuria and decreased survival for EP4pod+ mice while EP4pod-/- mice were protected against the development of glomerular injury. Furthermore, our findings were supported by in vitro studies using cultured mouse podocytes where an adhesion defect was uncovered for cells overexpressing the EP4 receptor. Additionally, our investigations have demonstrated a novel synergy between angiotensin II AT1 receptors and prostaglandin E2 EP4 receptors. This was revealed by in vitro studies using isolated mouse glomeruli. There we were able to show that Ang II stimulation leads to increased expression of cyclooxygenase 2 (COX-2), the enzyme responsible for synthesis of PGE2, in a p38 mitogen activated protein kinase (MAPK) dependent fashion. Moreover increased PGE2 synthesis was measured in response to Ang II stimulation. We confirmed the presence of this synergy in our cultured mouse podocytes and showed an adhesion defect in response to Ang II stimulation which was COX-2 and EP4 dependent. These findings suggest that Ang II AT1 receptors and PGE2 EP4 receptors act in concert to exacerbate glomerulopathies. Studies using mice with either podocyte-specific overexpression of a dominant negative p38 MAPK or mice with global deletion of the EP1 receptor did not provide conclusive results as to their respective signaling involvement in podocyte injury. Altogether our findings provide novel insight for podocyte PGE2 EP4 and Ang II AT1 receptor signaling in models of CKD. These studies provide novel avenues for pursuing therapeutic interventions for individuals with progressive kidney disease.

Dopamine receptors belong to the G protein-coupled receptor (GPCR) superfamily and are classified into two families: D1-like (D1R and D5R) and D2-like (D2R, D3R and D4R), based on their ability to stimulate or inhibit adenylyl cyclase (AC). Classically, GPCRs (including D2R and D3R) are desensitized by the activation of the serine/threonine protein kinase C (PKC) upon phorbol-12-myristate-13-acetate (PMA) treatment. Previous studies demonstrate that while human D5R (hD5R) is also strongly desensitized upon PMA treatment, the human D1R (hD1R) undergo a robust PMA-induced sensitization. The aim of this PhD thesis was to explore how the canonical PKC- or phorbol ester-linked pathway can control the responsiveness of two similar GPCRs like hD1R and hD5R in an opposite fashion. Our data indicate that hD1R sensitization and hD5R desensitization are not mediated by a direct modulation of AC activity by PKC. Using a chimeric approach, we identified the third intracellular loop (IL3) as the key structural determinant controlling in an opposite manner the PMA-mediated regulation of hD1R and hD5R. To delineate the potential PKC phosphorylation sites, a series of mutation of serine (Ser) and threonine (Thr) located into IL3 of hD1R and hD5R were used. No hD1R mutation decreased the PMA-mediated sensitization. This suggests that hD1R phosphorylation is not required for PMA-induced sensitization. In contrast, our results indicate that PMA-mediated hD5R desensitization occurs through a hierarchical phosphorylation of Ser260, Ser261, Ser271 and Ser274. Notably, these hD5R mutants exhibited a PMA-induced sensitization, reminiscent of the PMA-induced hD1R sensitization. Additionally, using short hairpin RNAs (shRNAs), we showed that PKCε is the potentiating PKC while the desensitizing isoform is δ. Overall, our work suggests the presence or absence of specific Ser residues on IL3 of hD1-like receptors dictate if phosphorylation-dependent desensitization (through PKCδ) or phosphorylation-independent potentiation (via PKCε) will occur.

Background: RGS9-deficient mice show drug-induced dyskinesia but normal locomotor activity under unchallenged conditions. Results: Genes related to Ca2+ signaling and their functions were regulated in RGS9-deficient mice. Conclusion: Changes in Ca2+ signaling that compensate for RGS9 loss-of-function can explain the normal locomotor activity in RGS9-deficient mice under unchallenged conditions. Significance: Identified signaling components may represent novel targets in antidyskinetic therapy. The long splice variant of the regulator of G-protein signaling 9 (RGS9-2) is enriched in striatal medium spiny neurons and dampens dopamine D2 receptor signaling. Lack of RGS9-2 can promote while its overexpression prevents drug-induced dyskinesia. Other animal models of drug-induced dyskinesia rather pointed towards overactivity of dopamine receptor-mediated signaling. To evaluate changes in signaling pathways mRNA expression levels were determined and compared in wild-type and RGS9- deficient mice. Unexpectedly, expression levels of dopamine receptors were unchanged in RGS9-deficient mice, while several genes related to Ca2+ signaling and long-term depression were differentially expressed when compared to wild type animals. Detailed investigations at the protein level revealed hyperphosphorylation of DARPP32 at Thr34 and of ERK1/2 in striata of RGS9-deficient mice. Whole cell patch clamp recordings showed that spontaneous synaptic events are increased (frequency and size) in RGS9-deficient mice while long-term depression is reduced in acute brain slices. These changes are compatible with a Ca2+-induced potentiation of dopamine receptor signaling which may contribute to the drug-induced dyskinesia in RGS9-deficient mice.

Neonatal ventral hippocampus (nVH) lesioned animals show molecular and behavioral abnormalities analogous to those described in schizophrenia. As an extension to previous studies that showed an increase in ligand binding of cortical alpha-1 adrenergic receptors (AR) and a dysfunction in alpha-1 AR regulation of mesolimbic dopamine functions in post-pubertal nVH lesioned rats, we investigated the subcellular expression and activity of protein kinase C (PKC)---a second messenger in alpha-1 AR signaling---in the prefrontal cortex (PFC) and nucleus accumbens (NAcc) of post-pubertal nVH lesioned rats. Western blot analysis of membrane and cytosolic fractions showed complex changes in lesioned animals in the expression of different PKC subtypes following saline or alpha-1 AR agonist (cirazoline i.p.) injection. Among these changes, nVH lesioned animals showed a significant increase in membrane bound PKC alpha and phospho-PKC, and a decrease in cytosolic PKC gamma and PKC betaII in the PFC in comparison to sham-lesioned controls following saline. Cirazoline increased membrane bound PKC alpha in controls but decreased it in lesioned animals. In the NAcc, lesioned animals showed an increase in membrane bound and cytosolic PKC epsilon and PKC lambda levels following saline. Following cirazoline, lesioned animals showed a decrease in membrane bound PKC epsilon and PKC lambda, while controls showed an increase in cytosolic and membrane fractions of PKC epsilon with no change in PKC lambda. In vitro PKC activity assays showed increased basal activity in PFC slices of lesioned animals compared to controls, with no difference in NAcc slices. alpha-1 AR stimulation by the agonist phenylephrine (PE) increased PKC activity in PFC of controls while decreasing activity substantially in lesioned animals. In the NAcc, high concentrations of PE increased activity in controls, but decreased activity in lesioned animals. This abnormal expression and activity of PKC in the PFC and NAcc of nVH lesioned animals may be related to abnormal alpha-1 AR functions and may modulate some of the abnormal neuronal functions in these animals, such as working memory deficits and hyper neuronal excitability of the PFC and the NAcc.

The integrin family of cell surface receptors is evolutionary conserved and found in all multicellular animals. In humans 8-alpha and 18-beta integrins are non-covalently associated into 24 dimers. Integrins mediate cell-extracellular matrix and cell-cell interactions and participate in cell signalling. This ideally places integrins to regulate vital processes such as cell adhesion, migration, differentiation and cytoskeleton dynamics. Integrins also play a fundamental role in regulating cell survival and anoikis. In this thesis molecular mechanisms employed by integrins to induce signal transduction, independently or through crosstalk with other receptors, were characterised. Rictor-mTOR (mTORC2) was required for Akt Ser473 phosphorylation in response to β1 integrin-mediated adhesion as well as EGF-, PDGF- or LPA-stimulation of MCF7 cells. ILK and PAK were dispensable for Akt Ser473 phosphorylation upon β1 integrin-engagement or EGF-stimulation. PAK was needed when this phosphorylation was induced by PDGF or LPA. β1 integrin-promoted cell survival during serum starvation conditions was mTORC2 dependent, indicating the importance of Akt Ser473 phosphorylation. mTORC2 was also required for Akt Ser473 phosphorylation induced upon heparanase treatment of cells. Heparanase preferred PI3K catalytic subunit p110α for the upstream lipid phosphorylation required for Akt activation. Interaction between this subunit and Ras was needed for optimal Akt phosphorylation upon heparanase exposure. Cell adhesion strongly promoted heparanase signalling, which was more efficient in β1 integrin-expressing fibroblasts compared to cells lacking this subunit. The cooperative signalling between integrins and heparanase involved FAK and PYK2 since simultaneous silencing of these kinases suppressed heparanase-triggered Akt activation. Furthermore, the resistance of cells to apoptosis induced by H2O2 or serum starvation was promoted by heparanase.  Integrin stimulation by adhesion or cyclic stretching showed divergent downstream signalling responses. Cell attachment on integrin-specific ligands lead to robust phosphorylation of several intracellular integrin-effectors, e.g. p130CAS, FAK, Akt and ERK 1/2. However, mechanical cell stretching only triggered prominent phosphorylation of ERK 1/2. Signalling induced at early stages of integrin-mediated cell adhesion occurred independently of intracellular contraction. Reactive oxygen species (ROS) generated during adhesion and cell stretching influenced integrin signalling. Inhibition of mitochondrial ROS production blocked adhesion-induced Akt phosphorylation. In contrast, stretch-induced ERK 1/2 phosphorylation was elevated when extracellular ROS was scavenged. These results indicate that the two types of integrin stimuli generate signals by different mechanisms.