Fakultät für Biologie - Digitale Hochschulschriften der LMU - Teil 02/06

Elaboration of a dendritic arbor and the extension of an axon define the neuronal shape and are the key morphological features defining neuronal maturation. How the process is

molecularly regulated is only poorly understood.

In this study we used the dorsal multidendritic arborization (md-da) neurons of the Drosophila embryonic Peripheral Nervous System(PNS) to address the question about

the Roundabout (Robo) receptor protein function during dendrite field development. We identified Robo as one of the proteins involved in regulating the balance between dendritic branch elongation and new branch formation during dendritic arbor specification of the morphologically most complex, filling-in Class IV neuron of Drosophila PNS.

To dissect the role of the Robo proteins during dendritogenesis, we performed detailed

developmental analyses of dendrite field formation of md-da neurons and compared how dendrite morphogenesis differs in animals lacking Robo function or having too much of

the protein in their sensory neurons. We observed that changing the function of Robo protein results in defects in the number and elongation of high order dendritic branches.

With the help of cell-class specific genetic markers we also observed that Robo acts cell- class specifically and is required during dendrite field development of the

morphologically most complex class of md-da neurons, Class IV. Based on MARCM (mosaic analyses with a repressible cell marker) rescue and expression pattern experiments we suggest that this function of Robo is cell-autonomous. By doing time-lapse analyses we assessed the mechanistic role of overexpressing Robo during dendrite field development. We could verify that this protein limits the elongation and new branch formation of fine dendritic processes of the Dorsophila Class IV neurons. We performed expression pattern analyses, ectopic expression experiments and MARCM experiments for slit and suggest that muscles and neurons themselves are possible sources for the ligand. There are few molecules known to mediate the activation of Robo via Slit in the growth cone of an axon. Among these are Dock, a SH2-SH3 adaptor protein and Ena, a member of the VASP family of proteins. Loss-of-function analyses for these two proteins suggest that Ena is acting downstream of Robo during the regulation of dendrite field development of md-da neurons. Finally we propose a model in which Robo

responsiveness to Slit is down regulated via Robo2 during axon patterning of dorsal md- da neurons.

Der Hauptregulator der vaskulären Homöostase ist das Endothel, welches eine Vielzahl an vasoprotektiven Effekten ausübt. Die Integrität und Regulation des endothelialen Zellayers der Blutgefäße ist von großer Bedeutung bei physiologischen und pathologischen Prozessen. Die Basis dieser Phänomene ist in der dynamischen und exakt kontrollierten Regulation des Zytoskeletts begründet. Die wichtigsten Regulatoren des Zytoskeletts stellen die GTPasen der Rho-Familie und deren Effektoren dar.

Im Rahmen dieser Doktorarbeit untersuchten wir in primären humanen Nabelschnurendothelzellen, eine neu in Erscheinung tretende Zytoskelettstruktur, der wir in Anlehnung an ähnliche Proteingruppierungen in monozytären Zellen den Namen HUVEC-Podosomen gaben. HUVEC-Podosomen sind aufgrund ihrer Komponenten mit klassischen Podosomen vergleichbar. Allerdings gibt es zwischen beiden Strukturen auch Unterschiede, denn während klassische Podosomen aus Ring und Kern bestehen, zeigen HUVEC-Podosomen eine zweischichtige Architektur.

Wie wir weiterhin nachweisen konnten liegen Podosomen der Endothelzellen an der ventralen Plasmamembran und haben engen Kontakt mit der extrazellulären Matrix.. Somit fungieren sie, wie auch die klassischen Podosomen, als Adhäsionsstrukturen. Sie dienen aber nicht nur der Adhäsion, denn wie bei FITC-markierten Monolayer-Versuchen gezeigt werden konnte, haben sie auch eine proteolytische Aktivität, die insbesondere beim Matrixverdau und der daran anschließenden Migration von Bedeutung ist.

Ferner können wir zeigen, daß HUVEC-Podosomen in ruhenden, konfluenten Zellayern nicht beobachtet werden können. Sie lassen sich aber in migratorischen (subkonfluent oder nach Verwundung) HUVEC, in hoher Anzahl und diversen ringförmigen Formationen, vorwiegend in Bereichen nahe dem Leitsaum nachweisen. Wie wir mit Hilfe von live cell imaging-Experimenten zeigen konnten, sind diese Strukturen hochdynamisch und breiten sich wellenartig mit einem weiten Radius innerhalb einer Zelle aus. Scheinbar dispergieren diese Formationen oder fusionieren mit der Zellplasma, wodurch sie die enthaltenen Proteine für viele andere zytoplasmatische oder membranöse Prozesse freigeben könnten.

Durch Experimente, in denen Zytokine wie VEGF, bzw. Zytokin-produzierende Zellen wie Monozyten den HUVEC-Kulturen zugegeben wurden, konnten wir zeigen, daß diese die Bildung von Podosomen induzieren und sogar erheblich steigern. Unsere Arbeiten mit konstitutiv aktiven und dominant negativen GTPase-Mutanten zeigten weiterhin, daß diese bei der Organisation und Entstehung der HUVEC-Podosomen von entscheidender Bedeutung sind. Ferner konnte mit Hilfe von Mikroinjektionsversuchen von einer Teildomäne (A) des N-WASP-Proteins verifiziert werden, daß der Mechanismus zur Bildung der HUVEC-Podosomen eine Arp2/3-abhängige Aktinnukleation beinhaltet. Weiterhin ist die Bildung dieser Adhäsionsstrukturen auch von Src Tyrosinkinasen und PI3-Kinase abhängig.

Eine der Komponenten von HUVEC-Podosomen ist das Markerprotein Drebrin. Drebrin kann nur in diesen Strukturen und an Zell-Zell-Kontakten in HUVEC detektiert werden. Mikroinjektionsversuche von diversen Konstrukten der unterschiedlichen Regionen von Drebrin zeigen, daß dieses Protein von großer Bedeutung für die Bildung und Struktur der HUVEC-Podosomen ist. Die einzelnen Protein-Protein-Interaktionen von podosomalen Komponenten untereinander und mit Drebrin wurden mit Hilfe von Immunpräzipitation getestet. Es ist uns jedoch nicht gelungen einen Drebrin-Interaktionspartner zu finden. Eine Interaktion von Drebrin konnten wir nur mit Drebrin selbst in Form einer Dimerisierung bzw. mit F-Aktin nachgeweisen.

Es ist sehr wahrscheinlich, daß es sich bei HUVEC-Podosomen um ein multifunktionelles Organell handelt. Wie wir in dieser Arbeit darstellen, sind HUVEC-Podosomen Adhäsionsstrukturen. Sie können am häufigsten am Leitsaum detektiert werden, wobei ihre Generierung nur in Zellen mit migratorischen Phänotyp (in Zellen am Wundrand oder in s

Die Familie der Sorting Nexine (SNX) umfasst 33 bekannte Mitglieder, jedoch ist der Funktionsmechanismus vieler Sorting Nexine bislang nicht aufgeklärt. Auf der Suche neuer Modulatoren der βAPP-Proteolyse konnte im Rahmen eines Expressionsklonierungs-Screens (Schobel et al., 2006) ein bislang nicht beschriebenes Protein, Sorting Nexin 33 (SNX33), als Aktivator der βAPP-Proteolyse identifiziert werden.

SNX33 ist ein phosphoryliertes Protein, das ubiquitär exprimiert wird und zudem eine hohe Homologie zu den Proteinen SNX9 und SNX18 aufweist. SNX33 ist im Zytosol lokalisiert, kann jedoch auch Membran-assoziiert vorliegen. Es konnte gezeigt werden, dass Überexpression von SNX33 zu einer Inhibition Dynamin-abhängiger Endozytose und in Folge dessen zu einer etwa 50% -igen Reduktion der βAPP-Endozytose führt. Die von SNX33 induzierte Endozytosehemmung wird durch die SH3-Domäne des Proteins vermittelt. Im Rahmen dieser Doktorarbeit durchgeführte Koimmunpräzipitationsstudien zeigten, dass SNX33 mittels seiner SH3-Domäne mit Dynamin interagiert und auf diese Weise möglicherweise dessen Funktion moduliert. In Übereinstimmung mit den durchgeführten Zellkultur-Experimenten führte eine Überexpression von SNX33 im Modellorganismus Caenorhabditis elegans ebenfalls zu einem Dynamin-Funktionsverlust.

Da SNX33 Expression zu einer generellen Inhibition Dynamin-abhängiger Endozytose führt, handelt es sich dabei nicht um einen spezifischen βAPP-Modulator. Konsequenz einer reduzierten βAPP-Internalisierung ist eine starke Zunahme der neurotrophen sAPPα-Bildung sowie - je nach verwendeter Zelllinie - ein leichter Anstieg bzw. eine geringe Reduktion der pathogenen sAPPβ-Generierung. Es konnte gezeigt werden, dass Überexpression der homologen Proteine SNX9 und SNX18 ebenfalls zu einer Zunahme der βAPP-Spaltung führt. Es handelt sich also um einen Effekt, der von der ganzen Sorting Nexin-Subgruppe (SNX33/SNX9/SNX18) vermittelt wird. Diese Beobachtung legt die Vermutung nahe, dass diese Funktion innerhalb dieser Subgruppe konserviert ist. Transfektion von SNX1 führte zu keiner Änderung der βAPP-Proteolyse, was bedeutet, dass dieser Effekt nicht von der gesamten Sorting Nexin-Familie vermittelt wird. Interessanterweise ist die Spaltung von βAPP besonders sensitiv bezüglich einer veränderten Endozytose-Rate, da die Proteolyse der Transmembranproteine L-Selektin und des Tumornekrosisfaktor-Rezeptors 2 (TNFR2) unter SNX33 Überexpressionsbedingungen nicht signifikant verändert war.

Ein siRNA-vermittelter Knock-Down von SNX33 führte zu keiner generellen Endozytoseinhibition in HEK293 Zellen, es konnte keine veränderte βAPP-Endozytoserate beobachtet werden. Die Bildung von sAPPα- und sAPPβ war in Folge dessen unverändert. Auch ein lst-4/SNX33-Knock-Down in C. elegans führte überraschenderweise zu keiner Inhibition der Dynamin-Funktion, äußerte sich jedoch in einer Fehlfunktion der Insulin-Signaltransduktion. SNX33-Knock-Down in humanen Zellen brachte keine nachweisbare Beeinträchtigung des Insulinsignalweges mit sich, jedoch besteht die Möglichkeit, dass die Homologen SNX9 und SNX18 einen Verlust von SNX33 kompensieren können. Dabei gilt zu beachten, dass eine Funktionsübernahme durch homologe Proteine in C. elegans nicht möglich ist, da dieser Organismus nur ein einziges homologes Protein der SNX33/SNX9/SNX18-Subgruppe besitzt.

Im Rahmen dieser Doktorarbeit präsentierten sowie diskutierten Daten zeigen, dass SNX33 in unterschiedliche zellulärer Prozesse involviert ist. SNX33 ist ein neu identifizierter Modulator der Zelle, der für zentrale Signalwege und Vorgänge, wie zum Beispiel der Insulinrezeptor-Signaltransduktion und Endozytose, von Bedeutung ist. Im Gegensatz zum Modellorganismus C. elegans kann im humanen Zellkultursystem ein durch siRNA induzierter Funktionsverlust von SNX33 durch die homologen Proteine SNX9 und SNX18 kompensiert werden.

In spite of strong evidence that the mammalian cell nucleus is a highly organized organelle, a

consensus on basic principles of global nuclear architecture has not so far been achieved.

The existence of major architectural features such as an organized interchromatin

compartment and higher order organization of chromatin postulated by some of the models is

questioned or even refused by the others. This study was set up to test predictions of the

various model views after manipulating nuclear architecture by applying the induced

formation of hypercondensed chromatin (HCC). This method leads to massive but

completely reversible conformational changes of chromatin arrangements in living cell nuclei,

but does not affect the cells survivability.

Nuclear functions like transcription, replication and cell cycling were immediately stalled

when HCC formation was induced, but were rapidly recovered upon recovery of normal

chromatin configurations. The emerging pattern of HCC revealed a 3D network of

interconnected chromosome territories. The surface of the emerging HCC bundles was the

site of preceding activity like RNA transcription or DNA replication, which confirmed the

existence of a distinct topological arrangement of functional processes with respect to the

architecture of chromatin. This arrangement could further be demonstrated by analyzing the

topography of defined chromatin modifications, showing that active chromatin is preferentially

located at the HCC bundle surfaces, whereas inactive chromatin regions are preferentially

found in the HCC bundle interior. The emerging patterns of HCC were further strikingly

similar in consecutively repeated cycles of HCC formation and recovery, demonstrating a

non-random but pre-existing and defined chromatin and interchromatin topography. All

results of this study were obtained using confocal laser scanning microscopy. A protocol for

deconvolution of confocal images was established to enhance confocal image quality to an

extent sufficient for subsequent image analysis.

In contribution to the present model views this study demonstrates: [1] That most chromatin

exists in the form of higher-order sub-compartments ('~1 Mb chromatin domains') above the

level of extended 30 nm fibers and [2] That an interchromatin compartment exists as a

dynamic, structurally distinct nuclear compartment, which is functionally linked with the

chromatin compartment.

An updated chromosome territory-interchromatin compartment model on the basis of the

gained results is presented at the end of this thesis together with an attempt to provide a

comprehensive view linking ultrastructural with light microscopic insights.

Gene duplication is an opportunity for evolving new functions from the newer gene, but also has a disadvantage due to local gene-rearrangement effects and, if duplications are numerous, through alterations of genome size. Therefore, selection is playing a central role in determining the fate of a duplicate gene. Plants are known to harbor numerous gene families, and are thus an ideal system to test the fate of gene duplicates. This thesis tackles the tropinone-reductase like enzymes (further TRL) and the tau GSTs located upstream from this gene family. TRL enzymes are short-chain dehydrogenases that are involved in a reduction step downstream in the synthesis of tropane alkaloids in Solanaceae, important defense compounds of plants. The function of TRLs in Brassicaceae is not clear, since most of the plants in this family do not produce tropane alkaloids, but some have been associated with the oxidative-stress response. This gene family contains 80\% of its members duplicated in tandem in Arabidopsis thaliana. We profited from this fact to isolate 12 TRL (+ pseudogenes) from this species, further six species of Brassicaceae (A. thaliana, A. lyrata, A. cebennensis, Capsella rubella, Boechera divaricarpa and Brassica rapa), and one species from a closely related plant family, Cleome spinosa. We tested the role that selection plays in maintaining large numbers of this gene family. We used phylogenetic methods to analyze non-coding sequence evolution and identified regulatory motifs. We analyzed non-coding sequence evolution. Microarray expression data from A. thaliana and qPCR for A. thaliana and A. lyrata were analyzed to detect divergence in the expression patterns of orthologs and paralogs. TRL genes follow a gene birth and death dynamics. More probable, they originated from non-equal recombination of tandem duplicated genes. Positive selection at the origin of the duplicated genes allowed these to acquire differential expression patterns, leading to the preservation of numerous TRLs. The analysis of coding and non-coding sequences shows them to display correlated evolution, particularly in species recently separated by speciation. We further tested for selection on the tau glutathione-S-transferases (GST) enzymes, adjacent 3' in the genome to TRLs. Tau GSTs are unique to plants and are involved in detoxification. Multiple copies of these enzymes will allow flexibility in substrate specificity, which is important for the detoxification function. We detected positive selection among paralogs of tau GSTs supporting their potential of functional diversity, but we also detected negative selection among paralogs and groups of orthologs, indicating that more often their functions are conserved.

This work is dedicated to studying natural variation in D. melanogaster at the DNA sequence and gene expression level. In addition I present a new version of the DNA polymorphism analysis program VariScan, which includes significant improvements.

In CHAPTER 1 I describe a genome scan of single nucleotide polymorphism in two natural D. melanogaster populations (from Africa and Europe) on the third chromosome. Together with polymorphism data previously published for the X chromosome of the same populations, this allows a comparative study of the polymorphism patterns of the X chromosome and an autosome. The frequency spectrum of mutations and the patterns of linkage disequilibrium are investigated. The observed patterns indicate that there is a significant difference in the behavior of the two chromosomes, as has already been suggested by previous studies. To uncover the reasons for this a coalescent based maximum likelihood method is applied that incorporates the effects of demographic history and unequal sex ratios. For the African population the differential behavior of the chromosomes can be explained by its demographic history and an excess of females. In Europe, a population bottleneck and an excess of males alone cannot explain the patterns we observe. The additional action of positive selection in this population is proposed as a possible explanation.

In CHAPTER 2 I investigate the variation in gene expression of the two aforementioned populations. Whole-genome microarrays are used to study levels of expression for 88% of all known genes in eight adult males from both populations. The observed levels of expression variation are equal in Africa and Europe, despite the fact that DNA sequence variation is much higher in Africa. This is evidence for the action of stabilizing selection governing levels of expression polymorphism. Supporting this view, genes involved in many different functions, and are therefore on strong selective constraint, show less variation than do genes with only few functions. The experimental design allows the search for genes which differ in their expression patterns between Europe and Africa and might therefore have undergone adaptive evolution. Detected candidates include genes putatively involved in insecticide resistance and food choice. Surprisingly, many genes over-expressed in Africa are involved in the formation and function of the flying apparatus.

In CHAPTER 3 I present version 2 of the program VariScan. This program was designed to analyse patterns of DNA sequence polymorphism on a chromosomal scale. The functionality of the core analysis tool, the wavelet decomposition, is described. In addition, multiple improvements to the previous version are presented. The program now supports the “pairwise deletion” option. This is essential for analysing data at the chromosome scale, since such data often contains incomplete information. It is now possible to add outgroup information, which allows the calculation of additional statistics. Furthermore, the separate analysis of different predefined chromosomal regions is added as an option. To increase the user friendliness, a graphical user interface is now included as part of the software package. Finally, VariScan is applied to published and computer-generated data and the ability of the wavelet-based analysis to uncover chromosomal regions with interesting DNA polymorphism patterns is demonstrated.

Mitosis is the process by which sister chromatids are equally segregated into two daughter cells. Tight control in various events during mitotic progression is essential for maintaining chromosome stability. Mitotic kinases including Cyclin dependent kinase 1 (Cdk1) and Aurora family are required for regulating proper mitotic progression by phosphorylating mitotic substrates thereby, controlling their activities, localization or abundance. On the other hand, these mitotic kinases are modulated by de-novo synthesis, activators, phosphorylation and ubiquitin-dependent proteolysis. A thorough understanding of the function and regulation of mitotic kinases could further our knowledge on mitotic progression.

In the first part of the thesis, we investigated the expression, localization and regulation of human Lats1 kinase, which is a close homologue of the yeast Dbf2 kinase family involved in the mitotic exit network (MEN). Despite the fact that Lats1 has been suggested to be a spindle protein that binds and inactivates Cdk1, we found that Lats1 is mainly cytoplasmic throughout the cell cycle by immunofluorescence microscopy. Both yeast two-hybrid and coimmunoprecipitation showed no significant interaction between Lats1 and Cdk1. Although Lats1 was highly phosphorylated during mitosis, no detectable kinase activity was observed. However, we identified Ste20 like kinase MST2 as the upstream regulator of human Lats1. Phosphorylation of Lats1 by Mst2 resulted in the activation of Lats1 kinase activity both in vivo and in vitro. This kinase-substrate relation was proven to be specific, as another distant Mst2 homolog, Mst4, did not possess this ability. Subsequent mass-spectrometry-based phosphosites analysis revealed that Mst2 phosphorylates Lats1 on more than five residues. Alanine mutations on Lats1T1079 and S909 impaired Lats1 kinase activity. Thus, we could not confirm the suggested role of Lat1 in mitosis. Instead, we show that similar to its Drosophila ortholog, Lats1 is involved in the Mst2 signaling pathway and might control developmentally regulated cell proliferation and apoptosis in mammals.

In the second part of this thesis, we characterized hBora, a novel Aurora A interactor originally found in Drosophila. We show that hBora is upregulated and phosphorylated during mitosis. siRNA-mediated knockdown of hBora led to spindle formation defects and aneuploidy. hBora overexpression caused monoastral spindle formation and mislocalization not only of Aurora A but also Plk1. Further investigations showed that Cdk1 phosphorylation on hBoraSer252 leads to Plk1 binding and this may promote the SCF-mediated proteolysis of hBora. Indeed, Plk1 depletion led to an increase in hBora levels. Interestingly, the co-depletion of both hBora and Plk1 (to lower hBora levels in Plk1 depleted cells) rescued the localization of Aurora A to the centrosomes and bipolar spindle formation. Thus, we propose that hBora is a functional link between Plk1 and Aurora A and that by modulating the proteolysis of hBora, Plk1 could regulate Aurora A localization and activity. At the end, we also investigated the function of Aurora A and could show that Aurora A is required for centriole cohesion and centrosome separation.