Diss. (sammanfattning) Stockholm : Univ., 2004. / Härtill 3 uppsatser.

Thesis (Ph. D.)--University of Washington, 1998. / Vita. Includes bibliographical references (leaves [47]-52).

No description available.

No description available.

Musashi-2 (MSI2), a member of the Musashi family of RNA-binding proteins, is thought to play a critical role in the maintenance of stem cell populations and in the formation of aggressive tumours. Multiple studies indicate that MSI2 plays an important role in the maintenance of hematopoietic stem cell (HSC) populations and recent studies in humans identify MSI2 as an independent prognostic factor for overall survival in patients with Acute Myeloid Leukemia (AML). Importantly, though correlative studies implicate MSI2 as a contributor to aggressive disease in human AML, no study to date has attempted to analyze the functional role of MSI2 in primary human AML samples. Furthermore, though MSI2 is critical for the maintenance of HSCs, the mechanisms through which MSI2 functions are unknown. The work presented in this thesis elucidates the biochemical mechanisms through which MSI2 functions and examines the functional role of MSI2 in human AML. Using a lentiviral-mediated shRNA knockdown of MSI2, I demonstrate that MSI2 is critical for the maintenance of human AML. A loss of MSI2 greatly impairs the ability of AML samples to maintain disease in a xenotransplantation assay. MSI2 is an RNA binding protein that is thought to repress the translation of target mRNAs in the cytoplasm and prevent the maturation of microRNAs (miRNAs) in the nucleus. The targets of MSI2 are believed to be potent regulators of stem-ness and dysregulation of these targets could very well contribute to neoplastic transformation. Cross-linking immunoprecipitation followed by next generation sequencing (CLIP-Seq), revealed the RNA binding properties of MSI2 and the RNA targets bound by MSI2. To identify novel MSI2 protein interactors, the MSI2 locus was endogenously tagged with the promiscuous biotin ligase BirA* and subjected to BioID analysis. When compared to appropriate controls, we were able to robustly identify proteins that associate with MSI2. The analysis of one of these protein binding partners, Insulin-like growth factor 2 mRNA binding protein 2 (IGF2BP2) reveals a critical role in the normal function of HSCs. / Thesis / Doctor of Philosophy (PhD) / The hematopoietic system is responsible for the production of billions of mature cells everyday. These mature cells are “differentiated”, meaning that they have gone through a process that has allowed them to become specialized to perform a very specific role. Throughout the process of differentiation, most functional cells lose their ability to proliferate. The continued production of these functional cells comes from a pool of rare, quiescent, hematopoietic stem cells (HSC). These cells maintain the production of mature cells throughout the lifetime of an organism. The Musashi-2 (MSI2) protein has been identified as a protein that is critical for the normal function of HSCs. By altering the levels of the MSI2, it is possible to greatly impair or enhance the activity of HSCs. Moreover, correlative studies implicate MSI2 as a contributor to aggressive Acute Myeloid Leukemia (AML), a disease that occurs when HSCs become dysregulated. Despite its important roles in normal and abnormal hematopoiesis, very little is known about how MSI2 functions and whether it actually has a functional role in AML. We set forth to identify mechanisms through which the MSI2 protein functions and to prove that MSI2 contributes to the maintenance of human AML. We reveal that the MSI2 protein plays a critical role for the maintenance of human AML and identify novel pathways through which the protein functions. Importantly, MSI2 is known to interact with mRNA in order to alter post-transcriptional gene expression. We thoroughly characterize the RNA-binding characteristics of MSI2 and identify a plethora of MSI2 RNA targets. In an unbiased manner, we also identify a list of MSI2-protein interactors. We identify one MSI2 protein-binding partner, Insulin-like growth factor 2 mRNA binding protein 2 (IGF2BP2) that is preferentially expressed in the most immature fraction of HSCs and is critical for the proper function of HSCs.

The AID/APOBEC protein family comprises a group of cytosine deaminases found in vertebrates that are capable of modifying cytosine to uracil in the context of RNA or singlestranded DNA. They exert diverse valuable physiological functions including antibody diversification and restriction of viral infection. However, off-target mutations have also been shown to contribute to cancer development, making it crucial to better understand the interactions and mechanisms that regulate AID/APOBEC activity and editing site fidelity. In this regard, a new focus on RNA as a putative regulator of AID/APOBECs has recently emerged. Regardless of whether it is used or not as a substrate for deamination, most members of the family have been shown to retain the ability to bind RNA, emphasizing a potential regulatory role for this interaction. However, little is known about AID/APOBECs RNA binding specificity. A promiscuous binding has been suggested in some cases while in vitro evidence for other members of the family indicate a certain level of specificity. Therefore, to thoroughly unravel the AID/APOBECs RNA binding specificity, in my doctoral research I applied cross-linking and immunoprecipitation (iCLIP), an unbiased technique that allows identification of protein-bound RNAs with nucleotide resolution in living cells. As a first approach, I adapted the technique for its use in yeast and probed the RNA binding of AID and APOBEC3G, revealing different degrees of preference for small structured RNAs and recognition of particular sites within them. I then expanded the analysis to mammalian cells (HEK293T) and evaluated an extended set of APOBECs finding that, even in the presence of a broader and more complex pool of RNAs, small RNAs were still significantly bound by some members of the family. Furthermore, the comparative analysis of AID, APOBEC1, APOBEC3G, APOBEC3A and APOBEC3B iCLIP data obtained in my research, revealed shared and individual preferences for certain RNAs, suggesting a degree of binding specificity among APOBECs. In summary, my thesis outlines for the first time a comprehensive analysis of the RNA binding specificity of different AID/APOBECs in vivo, including the description of novel interactions with nucleotide resolution. The results obtained are of great value and open the field for further investigation of the specific meaning and validation of each preferential binding, providing new insights into understanding the role of AID/APOBEC interaction with RNA.

Evolutionarily conserved Pumilio (Pum) RNA-binding proteins act as translational repressors during embryo development and cell fate specification. Previous work in the lab has shown that over-expression of Pum2 (Pum2-EML) supports maintenance and suppresses mutilineage differentiation of murine multipotent HSC/MPP-like cell line EML. The subsequent analysis of HSC markers and functional analysis has revealed that wt EML cells share the LKS CD34 positive phenotype, whereas the majority Pum2-EML cells are similar to LKS CD34 negative. The CD34 positive wt EML cells can be divided into CD34low, CD34med and CD34high subpopulations, whereas Pum2-EML CD34 positive cells correspond to CD34low subpopulation. Colony forming assays have revealed that the overall multilineage differentiation of wt EML and Pum2-EML cells strongly correlates with the CD34 expression levels. Multiple experiments have revealed that purified CD34 negative and CD34 positive wt EML cells can generate each other and among CD34 positive wt EML cells the CD34low cells have the highest capacity to give rise to CD34 negative EML cells. We have proposed a model in which CD34 negative EML cells are more primitive cells in an "inactive" (differentiation inhibited) state, that give rise to CD3low "active" (differentiation ready) EML cells. The CD34low EML cells can revert back to the CD34 negative state or give rise to CD34med/high cells that can readily differentiate into multiple lineages. Based on that model, the over-expression of Pum2 leads to increased maintenance of cells in inactive CD34 negative state, and blocks development of CD34 positive cells past the CD34low stage. Cumulatively, these results support the notions that Pum2 could be involved in maintaining the balance between inactive and active state of multipotent hematopoietic cells. The c-kit receptor plays a vital role in self-renewal and differentiation of hematopoietic stem cells (HSC) and multipotent progenitors (MPPs). We have discovered that besides c-kit, the murine multipotent HSC/MPP-like cell line EML expresses the transcript and protein for a truncated form of c-kit, called tr-kit. Notably, the tr-kit transcript and protein levels were down-regulated during cytokine induced differentiation of HSC/MPP-like cell line EML into myelo-erythroid lineages. RT-PCR results show tr-kit is transcribed solely in cell populations enriched for LTR-HSC, STR-HSC and MPPs. The observation that tr-kit is co-expressed with c-kit only in more primitive, HSC and MPP-enriched cell populations raises an exciting possibility that tr-kit functions either as a new component of SCF/c-kit pathway, or is involved in a novel signaling pathway, present exclusively in HSC and MPPs. These findings necessitate functional characterization of tr-kit, and analysis of its potential role in the self-renewal, proliferation and/or differentiation of HSC and multipotent progenitors.

The RNA-binding protein (RBP) and Hu/ELAV family member HuD regulates mRNA metabolism of genes that encode proteins involved in neuronal differentiation, learning and memory, and certain neurological diseases. Given the important functions of HuD in a variety of processes, we set out to characterize the 5’ genomic region of the mammalian HuD gene and determine the mechanisms that regulate its mRNA expression in neurons using P19 cells and mouse brain as models. Bioinformatic and 5’RACE (rapid amplification of cDNA ends) analyses of the HuD 5’ genomic flanking region identified eight conserved leader exons (E1s), two of which are novel. Expression of all E1 variants was established in differentiating P19 cells, mouse embryonic (E14.5) and adult brains. Through several complementary approaches, we determined that the abundance of HuD mRNA is predominantly under transcriptional control in differentiating neurons. Sequential deletion of the 5’ regulatory region upstream of the predominantly expressed E1c variant revealed a well-conserved 400 bp DNA region that contains five E-boxes and is capable of directing expression of HuD specifically in neurons. Using electrophoretic mobility shift assays (EMSAs), chromatin immunoprecipitations (ChIPs), and E1c 5’ regulatory region (RR) deletion and mutation analysis, we found that two of these E-boxes are targeted by neurogenin 2 (NGN2/NEUROG2) and that this mechanism is important for induction of HuD mRNA in neurons. Additional deletion and mutation of the E1c 5’ RR revealed that putative cis-acting elements for Kruppel-like factors (KLFs) and nuclear DEAF-1-related (NuDR) transcription factors also positively regulate transcription of HuD. Together, our findings reveal that the intricate transcriptional regulation of mammalian HuD involves eight leader exons and potentially alternate promoters. We further demonstrate that transcription of HuD requires neuron-specific control by NGN2 and possibly KLF and NuDR transcription factors. To our knowledge, this is the first study to identify transcriptional events that positively regulate expression of HuD.

The RNA-binding protein (RBP) and Hu/ELAV family member HuD regulates mRNA metabolism of genes that encode proteins involved in neuronal differentiation, learning and memory, and certain neurological diseases. Given the important functions of HuD in a variety of processes, we set out to characterize the 5’ genomic region of the mammalian HuD gene and determine the mechanisms that regulate its mRNA expression in neurons using P19 cells and mouse brain as models. Bioinformatic and 5’RACE (rapid amplification of cDNA ends) analyses of the HuD 5’ genomic flanking region identified eight conserved leader exons (E1s), two of which are novel. Expression of all E1 variants was established in differentiating P19 cells, mouse embryonic (E14.5) and adult brains. Through several complementary approaches, we determined that the abundance of HuD mRNA is predominantly under transcriptional control in differentiating neurons. Sequential deletion of the 5’ regulatory region upstream of the predominantly expressed E1c variant revealed a well-conserved 400 bp DNA region that contains five E-boxes and is capable of directing expression of HuD specifically in neurons. Using electrophoretic mobility shift assays (EMSAs), chromatin immunoprecipitations (ChIPs), and E1c 5’ regulatory region (RR) deletion and mutation analysis, we found that two of these E-boxes are targeted by neurogenin 2 (NGN2/NEUROG2) and that this mechanism is important for induction of HuD mRNA in neurons. Additional deletion and mutation of the E1c 5’ RR revealed that putative cis-acting elements for Kruppel-like factors (KLFs) and nuclear DEAF-1-related (NuDR) transcription factors also positively regulate transcription of HuD. Together, our findings reveal that the intricate transcriptional regulation of mammalian HuD involves eight leader exons and potentially alternate promoters. We further demonstrate that transcription of HuD requires neuron-specific control by NGN2 and possibly KLF and NuDR transcription factors. To our knowledge, this is the first study to identify transcriptional events that positively regulate expression of HuD.

Oskar (Osk) is a posterior body patterning determinant in Drosophila melanogaster oocytes. oskar (osk) mRNA is translationally repressed until it reaches the posterior of the oocyte where Osk protein accumulates. Translational repression of osk prior to posterior localization is mediated by the RNA binding protein, Bruno (Bru). To better define Bru binding sites, I performed in vitro selections using full length Bru and the fragments containing either the first two RRMs (RRM1+2) or the third RRM (RRM3+). The aptamers from the final round from each of the selections produced a multitude of overrepresented primary sequence motifs. Examples of each of these motifs were found in the 3’UTRs of the mRNAs that Bru is known to regulate during oogenesis. GFP reporter transgenes under the control of the UAS-Gal4 expression system were constructed with each class of the binding sites within the reporter transgenes’ 3’UTRs to test the motifs’ ability to repress the reporters in vivo. In a wildtype background, the GFP reporters containing the binding sites were translationally repressed. In the aret mutant background, the GFP levels of the repressed GFP reporters increased with reduced Bru activity, suggesting the transgenes’ repression is mediated by Bru. Three of the motifs isolated in the in vitro selections reside in the AB and C regions of the osk 3’UTR, and the three classes of sites were mutated in the AB and C regions. The mutated AB and C regions were used to assay for a reduction of Bru binding affinity for the mutant RNAs. Additionally, the mutations were incorporated into an osk genomic transgene that was introduced into an osk RNA null as well as an Osk protein null background. The mutations reduced Bru binding to the AB and C regions. The transgenes containing the mutated Bru binding sites could not fully rescue the osk RNA null phenotype but can fully rescue the Osk protein null phenotype, suggesting an osk transcript can regulate other osk mRNAs in trans. / text