Border cells, a group of specialized follicle cells that commit collective migration during the oogenesis of Drosophila, constitute a useful migration model. Previous work in our laboratory by Kalman Somogyi identified Mal-D, a transcriptional co-activator of DSRF, is important for border cell migration. mal-D mutation causes decrease of F-Actin levels and loss of cellular integrity in border cells. Moreover Mal-D was found to accumulate in the nucleus of some border cells while the cluster is migrating and only if the cluster is migrating. A suggested mechanism was that the border cells receive a migration related signal, such as an increase of cellular tension and send Mal-D to the nucleus. The first part of my project was to understand how Mal-D is regulated by the migration. In order to visualize subcellular distribution of Mal-D I generated a tagged version of the endogenous protein by using homologous recombination. Analysis of subcellular distribution of Mal-D with this tool showed that the increase in nuclear levels of Mal-D in migrating cells is the result of an overall increase in the level of Mal-D protein and not redistribution of a fixed amount of protein. Furthermore I identified that mutations in Profilin or DSRF affect the nuclear levels of Mal-D. In the second part of my project I focused on the targets of Mal-D. I isolated border cell mutant for Mal-D or wild-type, and I compared their gene expression profiles by using microarrays.
The microtubule (MT) cytoskeleton is important for establishing polar growth in the rod-shaped fission yeast (Schizosaccharomyces pombe). In these cells, MTs form an architectural scaffold of the cell by positioning organelles such as the nucleus and mitochondria. Interphase MTs are arranged in bundles along the cell’s long axis. The filaments start growing in the cell’s middle in a zone of anti-parallel overlap, from which the more dynamic plus ends of MTs extend towards both cell ends. After cell division the cell grows exclusively from the old end (away from the septum), where the growth machinery is still present from the mother cell. New end take off (NETO) occurs after about a third of the way through the cell cycle, when F-actin has moved into the new end. From this point onwards maintenance of polar growth is MT independent and occurs at both cell ends. Guidance of the microtubules to the cell ends is performed by plus end tracking proteins (+TIPs), such as Tea1 and Tip1 (Clip-170). Tea1 is a landmark protein localizing to the cell ends. Tip1 is an anti-catastrophe factor that prevents MT depolymerization before the filament has reached the cell end. The delivery of Tip1 to MT ends is motors dependent and another +TIP, Mal3, anchors it at the MT end. Mal3 (EB1) stabilizes MTs, possibly by fortify its seem. Here we describe a large-scale, electron tomography investigation of wild-type (WT) S. pombe cells, including the first 3D reconstruction of a complete eukaryotic cell volume. Sufficient resolution to show both how many MTs there are in a bundle and their detailed architecture was achieved. Most cytoplasmic MTs are open at one end and capped at the other, providing evidence about their polarity. Electron-dense bridges between the MTs themselves and between MTs and the nuclear envelope were frequently observed. Finally, we have investigated structure/function relationships between MTs and both mitochondria and vesicles. Using the same approach, we then analyzed the bundle architechture in tip1Δ and mal3Δ mutants. MTs were half the length of WT in mal3Δ and a quarter the length of WT in tip1Δ. Further, there were less than half as many MTs in a bundle in tip1Δ then in WT. In contrast, mal3Δ bundles no difference in the amount of filaments in a bundle. However, structural differences of the MT lattice were observed in both mutants. The interaction between MTs and the spindle pole body was altered in both strains. Our analysis shows that electron tomography of well-preserved cells is ideally suited for describing fine ultrastructural details that were not visible with previous techniques.
Cancer is a dynamic process that requires the stepwise deregulation of mechanisms affecting various cellular traits. During my PhD, I have characterized how the stress-activated p38alpha MAPK signaling pathway regulates the processes of cellular migration, proliferation, and survival in the context of oncogene-induced malignant transformation, which recapitulates the mechanisms of cancer initiation at the cellular level. My studies have been mostly based on the use of human and mouse cultured cells, which I have analyzed using both biochemical and cell biological approaches. In particular, the development of p38alpha-deficient cell lines and the application of retrovirally-based gene expression techniques have been very useful. The implementation of tools to measure the intracellular levels of reactive oxygen species (ROS) within living cells has also been key for my work. I have found that p38alpha regulates the process of malignant transformation at various levels. First, p38alpha negatively regulates cell cycle progression induced by mitogenic signals in both exponentially proliferating and confluent cells. Oncogene-expressing cells proliferate faster in the absence of p38alpha, which may be accounted for by the negative effect of p38alpha on cyclin D1 expression. Similarly, p38alpha controls the process of cell-cell contact-inhibition, which requires p27Kip1 accumulation and triggers G1-phase cell cycle arrest upon cell confluence. The process of contact inhibition is likely to involve uncharacterized membrane-associated signaling events. Accordingly, I have found that p38alpha regulates the membrane composition of oncogene-transformed cells. In addition to its negative role in cell proliferation, I have shown that p38alpha can interfere with the process of malignant transformation by sensing oxidative stress and inducing apoptosis. Thus, p38alpha becomes activated when oncogene-expressing cells accumulate high levels of carcinogenic ROS and, in turn, induces the elimination of the transformed cells by apoptosis. Interestingly, I have found that human cancer cell lines that contain high ROS levels have developed a mechanism to by-pass this p38alpha function. Finally, in contrast to its anti-proliferative and pro-apoptotic roles, I have found that p38alpha is an important mediator of cytokine-induce cell migration, a process that is thought to be important for cancer cell metastasis.
Nonsense-mediated mRNA decay (NMD) is a molecular pathway of mRNA surveillance that ensures the rapid degradation of mRNAs containing premature translation termination codons in all studied eukaryotes. Originally, NMD was thought of as a quality control pathway that targets non-functional mRNAs arising from mutations and splicing errors. More recently, NMD has been shown to also regulate normal gene expression and NMD thus emerged as one of the key post-transcriptional mechanisms of gene regulation. Despite the progress in the understanding of the role and mechanism of this pathway, the physiological impact of NMD on humans is not yet fully uncovered. To explore the functions of NMD in humans, I combined RNAi against the essential NMD factor UPF1 with genome-wide microarray analysis. My research indicate that NMD affects the expression of a large number of genes implicated in a wide diversity of functions although a majority seems to be affected indirectly and – consequently – do not represent legitimate NMD targets. The validation of five bona fide NMD transcripts allowed me to develop an assay to quantitate differences in NMD efficiency. Using three different strains of HeLa cells as a simple model, I have systematically analysed the molecular mechanism underlying quantitative differences in NMD efficiency. The results of this analysis show that the quantitative differences in NMD efficiency represent a stable characteristic of the investigated strains. Low NMD efficiency is shown to be functionally related to the reduced abundance of the exon junction component RNPS1 in one of the analysed HeLa strains. Furthermore, restoration of functional RNPS1 expression, but not of NMD-inactive mutant proteins, also restores efficient NMD in the RNPS1 deficient cell line. I conclude that cellular concentrations of RNPS1 can modify NMD efficiency and propose that the cell type specific co-factor availability represents a novel principle that quantitatively controls NMD. I also tested the hypothesis of NMD as a genetic modifier in the phenotypic expression of disease. A specific -thalassemia – common in Mediterranean Asia – was assayed as a model. My results do not support a role of NMD for the variable severity of this specific mutation leading to anemia.
The transcription factor C/EBPbeta gene has CREB responsive elements (CRE) in its promoter, and its transcription is regulated by CREB during adipogenesis. We have generated a mouse line with a deletion of the CRE elements on the C/EBPbeta promoter and stusied the role of these elements in macrophages. We show that the CREs are important for the induction of C/EBPbeta expression following treatment of the macrophages with IFNgamma/LPS. Moreover, we found two novel targets for C/EBPbeta transcription in macrophages, that are macrophage scavenger receptor 1 (Msr1) and interleukin 13 receptor alpha1 (IL13a1). We also show that the well-known regulation of the arginase 1 gene by C/EBPbeta is dependent on the ability of CREB to upregulate C/EBPbeta. FACS analyses on our bone marrow-derived macrophage population, showed that the cells are Mac1(+), F4/80(+) and Gr1(+), typical markers of Natural Suppressor macrophages. Taken together, the C/EBPbeta target genes found in the macrophage and the cell surface markers, suggest an immunosuppressive phenotype. we propose a novel role for C/EBPbeta in mediating the molecular switch fron inflammatory to immunosuppressive macrophages. In a separate project, we study the role of the Thr188 and Ser176, Ser180 and Ser184 phosphorylation sites, which are located in the regulatory domain of the C/EBPbeta protein. Thr188 is a known MAPK phosphorylation site, whereas the three serines, whether all or some, were recently shown to be targets for GSK3beta phosphorylation. We created two mouse lines in which either Thr188, or the three serines were mutated to alanines. We analyzed the expression of the mutant C/EBPbeta in various tissues, as well as the expression of C/EBPbeta target genes in primary macrophages from both the mouse lines. We found that the three serines have a role in modulating C/EBPbeta's autoregulatory loop as well as in reducing the transcription factor's transactivational activity. Moreover, based on the migration pattern of the mutant C/EBPbeta proteins, we propose a modl suggesting cooperativity between the MAPK and GSK3beta phosphorylation sites. We conclude that the phosphorylation sites in question are implicated, whether directly or indirectly, in the modulation of the transcription factor's activity.
In the fission yeast Schizosaccharomyces pombe, interphase microtubules switch to depolymerization almost exclusively at the poles of the cylindrical cell. Proper localization of these events is crucial for morphogenesis, and is thought to depend on factors preferentially localized at the cell poles. Using computer simulations we analyzed five different models of how microtubule dynamic instability might be regulated in order to reproduce the organization of microtubules observed in vivo. To evaluate the simulations we compared their output to nine traits of interphase cells. Using the simulation we could show that the shape of fission yeast induces forces on microtubule plus ends specifically at the cell poles. We found that the effect of these forces on microtubule dynamics was sufficient to reproduce the nine traits without the requirement of pole-specific factors. In vivo experiments with mutant cells confirmed that cell shape is essential for the proper organization of microtubules in fission yeast which indicates that force might also be relevant in living cells. We furthermore applied the simulation to study the organization of microtubules in mutant cells that were deleted of the microtubule tip tracking proteins Mal3 and Tip1. This lead to the proposal of a novel hypothesis how Tip1 might function in vivo to assist force in the discrimination of the cell poles from central parts of the cell cortex.