Fakultät für Biologie - Digitale Hochschulschriften der LMU - Teil 01/06 Archiv

Die vorliegende Arbeit hatte zum Ziel, (i) die regulatorische Funktion bestimmter integraler Membranproteine im Dictyostelium Zytoskelett und (ii) die biochemische Interaktion einer Kinase eines infektiösen Bakteriums mit Aktin aus der Wirtszelle genauer zu analysieren. (i) Die ubiquitären Mitglieder der CD36/LIMPII-Familie sind integrale Membranproteine, die als Lipidrezeptoren und Zelladhäsionsproteine in der Plasmamembran oder - mit bisher unbekannter Funktion - in Membranen endosomaler Vesikel vorkommen. In Dictyostelium discoideum führte die Inaktivierung eines lysosomalen Membranproteins aus dieser Gruppe zur Suppression des Phänotyps einer Profilin-minus Mutante. Im Zuge der vollständigen Sequenzierung des D. discoideum-Genoms konnte festgestellt werden, daß es neben diesem DdLmpA noch die beiden weiteren homologen Proteine DdLmpB und DdLmpC gibt. Da der Mechanismus der Suppression des Profilin-minus Phänotyps ungeklärt ist, wurden die beiden Isoformen im Rahmen der vorliegenden Arbeit genauer charakterisiert. Sowohl für DdLmpB wie auch für DdLmpC konnte die familientypische Membran-Topologie einer Haarnadelstruktur nachgewiesen werden. Dabei weist die zentrale, lumenale Domäne beider Proteine zahlreiche Glykosylierungen auf. Durch Immunofluoreszenz und Saccharosegradienten wurde die Lokalisation der drei Isoformen an endolysosomalen Vesikeln nachgewiesen. Es stellte sich dabei heraus, daß die drei DdLmp-Proteine in unterschiedlichen Vesikelpopulationen auftraten. Auch “pulse-chase“-Experimente mit TRITC-Dextran und nachfolgender Markierung der Vesikel mit DdLmp-spezifischen Antikörpern ergaben unterscheidbare Zeitmuster für die Rekrutierung der Membranproteine in Vesikeln. Die für DdLmpA oft beobachtete Kolokalisation mit Makropinosomen konnte z.B. für DdLmpB und DdLmpC nur selten festgestellt werden. Nach zahlreichen Versuchen und der Konstruktion von verschiedenen Vektoren konnte am Ende der praktischen Arbeiten eine DdLmpB-minus Mutante im Wildtyp-Hintergrund isoliert werden. (ii) Im zweiten Teil der Arbeit wurde die in der Literatur beschriebene Interaktion zwischen Aktin und der Kinase YopO, die durch Yersinia enterocolitica als Effektorprotein in die Wirtszelle transloziert wird, biochemisch genauer untersucht. Es konnte festgestellt werden, daß G-Aktin und nicht F-Aktin für die Aktiverung der YopO-Kinase verantwortlich ist. Dabei tritt Nichtmuskel-Aktin im Vergleich zum Muskel-Aktin als ein deutlich besserer Aktivator von YopO auf. Obwohl die aktivierte Kinase in vivo das Aktin-Zytoskelett beeinflußt, ist Aktin offensichtlich kein Substrat von YopO. Mittels Fluoreszenzspektroskopie konnte gezeigt werden, daß sowohl die native Kinase YopO als auch das durch Punktmutation inaktivierte YopO K269A die Polymerisierungskinetik von Aktin behindern. Für eine mutmaßliche Aktin-Binderegion von 20 Aminosäuren aus dem C-terminalen Ende konnte hingegen kein Effekt beobachtet werden. Der Einfluß von aktinbindenden Proteinen, aktinmodifizierenden Substanzen und YopO-bindenden GTPasen auf die Aktivierung der Kinase durch Aktin deutet darauf hin, dass die Aktivität der Kinase in der Wirtszelle nicht nur durch Aktin alleine, sondern auch durch weitere Zytoskelett-Komponenten reguliert wird.

The aim of this PhD thesis was to examine the endophytic colonization behavior of Azospirillum brasilense and Herbaspirillum sp. N3 on wheat roots. The application of the FISH method using species specific phylogenetic oligonucleotide probes and GFP tagging facilitated the detection of a differential colonization behavior by the A. brasilense strains Sp7 and Sp245 on three different wheat varieties (Triticum aestivum). For this purpose a confocal laser scanning microscope (CLSM) was used, which enabled three-dimesional analysis of bacterial colonization of the root. Especially GFP tagged strains were well suited for this application, as there was no pretreatment or sectioning of the root sample necessary. Strain Sp7 was only located on the root surface of all wheat cultivars, whereas strain Sp245 was also found inter- and intracellulary in the outer root cortex layers. There was no recognizable connection between the growth stimulating effect of the inoculum (PGPR-effect) and the localization of the bacteria. The most pronounced PGPR-effect could be observed with the Brazilian wheat cultivar, which seemed to gain greatest benefit of its partnership with A. brasilense due to a certain adaptation to the inoculum. As the production of the auxin IAA (indole-3-acetic acid) plays a major role in stimulating plant growth, the expression of the key gene ipdC (indole-3-pyruvate decarboxylase) was examined. For this, several methods were tested to generate a fusion of the ipdC promoter with a gfp or rfp reporter gene. Constructing a translational promoter fusion with the gfp variant mut3 on plasmid level made expression analyses possible. With this method the promoter region of strain Sp7 located directly upstream of the ipdC start codon was found to differ only in a few bases from strain Sp245. But further upstream a region of about 150 bases was identified in strain Sp245, which was missing in strain Sp7. For Sp245 two different fusions were constructed, which contained the Sp245 promoter region homologous to strain Sp7 and the whole promoter region of Sp245, respectively. With these constructs the importance of the promoter region only present in strain Sp245 for control and intensity of ipdC expression in A. brasilense Sp245 could be demonstrated. Additionally an induction of the corresponding ipdC promoter fusions of Sp7 and Sp245 was achieved when adding phenylalanine or tyrosine. Total promoter activity was higher in strain Sp7 than in strain Sp245, and ipdC expression appeared to be subject to a stricter control in strain Sp245. These results were confirmed, when the strains containing the promoter fusions were used as reporters for ipdC expression on wheat roots. A demonstration of the induction of the ipdC promoter by root exudates in situ was possible. Finally, isolate N3 from surface sterilized wheat roots was characterized in detail. According to the 16S rDNA sequence data the isolate was phylogenetically allocated to the genus Herbaspirillum. But a subsequent DNS-DNS hybridization ruled out, that the strain belonged to any of the known Herbaspirillum species. Thus, the isolate, which might represent a new species, was named Herbaspirillum sp. N3. A specific, 16S rRNA targeted probe was constructed, which facilitated the examination of wheat roots colonization by this bacteria using FISH. Additionally the strain was GFP tagged to enable the detection in uncut root material. By this, an unequivocal demonstration of the endophytic colonization by Herbaspirillum sp. N3 mainly within the intercellular spaces was possible.

Two projects are described within the scope of this thesis. The first project characterizes the ubiquitin-specific protease UBP41 as a protein which upon overexpression causes apoptosis induction in several mammalian cancer cell lines. The second project investigates a possible involvement of the lysosomal cysteine proteases cathepsin-L and cathepsin-B in apoptosis pathways induced by distinct death stimuli, in particular by the tumor necrosis factor a (TNF-a). Therefore, both projects examine a possible regulation of apoptosis induction by proteases that are part of one of the two major systems of protein degradation. UBP41 as a protease with deubiquitylating activity is expected to play a role in the ubiquitin/proteasome system which is the major proteolytic apparatus for the degradation of cytosolic proteins. Cathepsins, on the other hand, are lysosomal proteases that participate in the breakdown of membrane-associated proteins and of extracellular proteins that are taken up by endocytosis. Both, the ubiquitin/proteasome system as well as lysosomal proteases have been previously implicated in the regulation and mediation of apoptosis, and it therefore appeared particularly attractive to further study the effect of UBP41 and cathepsins on cell death signalling in more detail.

Die α-Untereinheit des Phycoerythrocyanins ist ein Chromoprotein, das eine Rolle bei der Photosynthese von Mastigocladus laminosus, einem Cyanobakterium, spielt. Dieses Protein besitzt einen seltenen Gallenfarbstoff als Chromophor, der nie in freier Form gefunden werden konnte. Daneben ist das Chromoprotein zu einer ausgeprägten reversiblen Photochemie fähig, die der Photochemie des Phytochroms ähnelt. Es konnte ein Rekonstitutionssystem etabliert werden, mit dessen Hilfe die Chromophorbindung aufgeklärt werden konnte. Für die korrekte Bindung des Chromophors ist dabei neben dem Apoprotein und einem Gallenfarbstoff ein neuartiges Enzym, eine sog. Lyase-Isomerase notwendig. Anhand von Rekonstitutionen mit chemisch modifizierten Chromophoren konnte mit Hilfe molekularbiologischer, spektroskopischer und chromatographischer Methoden auch ein Modell für den Mechanismus der Photochemie des Chromoproteins entwickelt werden.

Besides its fully translocated form, the prion protein (PrP) can exist in two transmembrane forms (NtmPrP and CtmPrP), which span the lipid bilayer in either direction. Certain mutations in the membrane-spanning segment of PrP have been shown to increase synthesis of CtmPrP and result in neurodegeneration. One of these mutations is A117V, which is associated with Gerstmann-Sträussler-Scheinker syndrome (GSS). Sequence-analysis of patients showing neurological disorders revealed a new point-mutation (G114V) in the central region of PrP. The age of affected individuals is strikingly low; the clinical symptoms overlap with GSS as well as Creutzfeldt-Jakob disease. Due to their close vicinity within the transmembrane domain of PrP and misincorporation of valine in both cases, PrP G114V and PrP A117V were investigated in comparative studies. Both mutations caused a similar cellular phenotype strongly differing from wild type PrP. Characteristics included reduced expression of surface PrP as well as increased intracellular accumulation. Besides differences in the cellular localization, a slowed-down metabolism resulting in prolonged transit to the cell surface was detectable. Long-term expression of mutant PrP caused strong agglutination and decline of human neuroblastoma cells that might be due to the increased synthesis of transmembrane forms of mutant PrPs as verified using a cell-free topology assay. When proteasomal degradation was perturbed, PrP G114V and PrP A117V revealed enhanced proteinase K resistance and increased detergent insolubility. The close similarities displayed by both mutations point to a common pathological mechanism that differs from other prion diseases, thus arguing for a distinct class of disease.

The molecular basis of the Mohr-Tranebjaerg syndrome: A structural and functional analysis of the proteins DDP1 and Tim13 Mohr-Tranebjaerg syndrome is a mitochondrial disorder caused by a defects in the biogenesis of the human TIM23 translocase Tim8 and Tim13 of yeast belong to a family of evolutionary conserved zinc finger proteins that are organised in hetero-oligomeric complexes in the mitochondrial intermembrane space (IMS). The TIM8-13 complex assists the import of Tim23, the major component of the translocase for matrix-targeted proteins. Mutations in DDP1/TIMM8A, the gene encoding the human homolog of Tim8, cause the Mohr-Tranebjaerg syndrome (MTS), a progressive neurodegenerative disorder. This work shows that DDP1 and human Tim13 are zinc binding proteins which together form a 70 kDa complex in the intermembrane space of human mitochondria. Similar to yeast, the human DDP1-hTim13 complex facilitates import of yeast and human Tim23. It has been additionally analysed the structural and functional consequences of a MTS-missense mutation (C66W) directly affecting the conserved Cys4 metal binding motif. In this connection the C66W mutation impairs the ability to bind zinc. As a consequence, the mutated DDP1 loses its ability to assemble into a hetero-oligomeric complex with its partner protein human Tim13. Thus, it was suggested that an assembly defect of DDP1 is the molecular basis of Mohr-Tranebjaerg syndrome in patients carrying the C66W mutation.

The aim of this study was to reveal factors affiliated with morphological plasticity in Daphnia cucullata. This planktonic crustacean shows a seasonal variation in body shape that is commonly known as cyclomorphosis. Here I show that both predator-released chemicals and water-soluble “alarm” signals from wounded Daphnia can induce certain morphological features in Daphnia cucullata. Specifically, the inductive effect of the “alarm” signals is minute compared to the predator-exposed animals. Moreover, laboratory-generated small-scale turbulence was identified to be a proximate factor in cyclomorphosis. This special kind of turbulence is compared to turbulence evoked by the movement of other organisms. A distinct protective effect of the morphological traits against invertebrate predation was revealed as the likely ultimate factor of cyclomorphosis. Furthermore, a unique method of ultrasonic microscopy reveals “hidden” morphological plasticity in daphnids. The armor of predator-exposed animals is much stronger than that of those not exposed to predators. I developed a new SEM-preparation method to get detailed insight in ontogenesis in Daphnia. I detected morphological defenses in predator-chemical-induced Daphnia cucullata already in the last embryonic stage. Induction experiments on transgenerational effects in Daphnia cucullata show that defended mothers release offspring that are better defended than the descendants of unthreatened parents. Furthermore, daphnids could asses their actual predation risk by taking both predator and prey density into account. Thereafter the degree of a defense is adjusted to the individual hazard, thus saving unnecessary costs associated with the development of a particular defense. The combined results of this study indicate that cyclomorphosis in Daphnia cucullata has evolved as an adaptation to variable environments.

In this study a new Dictyostelium STE20-like protein kinase DST2 (Dictyostelium STE20-like kinase 2) was cloned and characterised. STE20 (Sterile 20) kinase was first identified in yeast as a pheromone-induced serine/threonine protein kinase that acts upstream of a MAP kinase cascade. Based on the domain structure, DST2 belongs to the GCK subfamily of STE20-like protein kinases, which include the mammalian STE20-like kinases (MST1/2/3), oxidant stress response kinase SOK-1, and DST1 in Dictyostelium discoideum which phosphorylates severin, a gelsolin-like F-actin fragmenting protein. DST2 was discovered by screening of the D. discoideum cDNA project database using DST1 as query. To confirm the existence of the DST2 gene and its expression, Southern, Northern and Western analyses of DST2 were carried out. It revealed that DST2 may have two copies in the Dictyostelium genome and that DST2 was expressed during all stages of D. discoideum development. In vitro kinase assays with bacterially expressed fusion protein of full length DST2 (aa461), the catalytic domain (aa287) and the regulatory domain (aa174) showed that autophosphorylation of DST2 occurson the regulatory domain and phosphorylates severin in the presence of a Mn2+ or Mg2+. Purified catalytic domain of PKA phosphorylated the regulatory domain of DST2 and caused an increase in the basal autophosphorylation activity of DST2, suggesting that PKA may be a potential upstream kinase of DST2 through the phosphorylation of its regulatory domain. To understand the function of the non-catalytic domain of DST2, three C-terminal truncation constructs (aa1-421, aa1-368 and aa1-326) were used in comparison to full length DST2 in in vitro kinase assays. Deletion of C-terminal regions revealed an inhibitory region amino acids 326-461 of DST2. Gel filtration chromatography showed that DST2 was eluted in a broad peak ranging from approximately 63 kDa to 400 kDa, suggesting that DST2 may exist in vivo as a monomer as well as a high molecular weight complex. The influence of phosphorylated and unphosphorylated severin on F-actin solutions was investigated using falling-ball viscometry and fluorescence spectroscopy. It turned out that phosphorylation by DST2 inhibits the F-actin fragmenting activity of severin, suggesting that DST2 may be directly involved in actin-cytoskeleton rearrangements.

Eph receptors and their membrane associated ephrin ligands mediate cell-cell repulsion to guide migrating cells and axons. A peculiarity of this signaling system is that both receptors and ligands can transduce signals into the cell resulting in a bidirectional signaling mode. An important step of ephrinB ligand ‘reverse signaling’ is the regulated tyrosine phosphorylation of the cytoplasmic domain, initiating docking sites for downstream signaling adaptors. Moreover, ephrinB ligands can signal by interactions with various PDZ-domain containing proteins. Using a broad range of in vitro assays the presented work demonstrates that upon binding to their cognate receptor, ephrinB ligands rapidly activate Src family kinases (SFKs) which subsequently phosphorylate ephrinB. Tyrosine phosphorylation appears to be a transient event that is downregulated by the tyrosine phosphatase PTP-BL, which interacts with ephrinB via one of its PDZ-domains. Studies on PTP-BL and another multiple PDZ domain containing protein GRIP (Glutamate Receptor Interacting Protein) revealed that PDZ interactions with ephrinB are also regulated by EphB receptor binding but, unlike tyrosine phosphorylation, these interactions are long lasting. These findings led to postulate a ‘switch model’ for ephrinB reverse signaling: Tyrosine dependent signaling appears to be a rapid and transient event which is later replaced by stable PDZ-dependent signaling. EphB ‘forward’ signaling as well as ephrinB ‘reverse’ signaling are important for axonal pathfinding and cell migration during development. Prior to their repellent effect on migrating cells and growth cones, Eph receptors form a high affinity complex with their ligands at sites of cell-cell contact. Therefore, mechanisms have to be in place that allow cells to detach from each other permitting retraction and withdrawal. To overcome this adhesive barrier, the ectodomain of ephrinA ligands is cleaved by metalloproteinases and shed upon receptor binding. Intrigued by the previous findings that activated ephrins cluster in cells, we hypothesized that these Eph/ephrin clusters undergo endocytosis. We developed immunofluorescence internalization and co-culture assays to study clustering and endocytosis at cell-cell contact sites. We established an experimental setup to perform fast time lapse imaging studies of cells expressing different fluorescently tagged proteins. Cell contact-induced ephrinB-EphB complexes are rapidly endocytosed during the retraction of cells and neuronal growth cones. Endocytosis occurs in a bidirectional manner, leading to internalized complexes of full length receptor and ligand, a yet rarely observed phenomenon. Signaling inactive mutants of EphB receptors and ephrinB ligands lead to a strong adhesion between cells. Endocytosis is sufficient to convert this adhesion into the detachment of cells. Bidirectional endocytosis is necessary to efficiently promote axon detachment during growth cone collapse mediated by ephrinB ligands. On the cell biological level, bidirectional endocytosis of two full length transmembrane (TM) proteins is a new phenomenon. Moreover, these studies reveal a novel mechanism of signal termination, de-adhesion and promotion of cell repulsion after intercellular (trans) interaction between two TM proteins.

Schalllokalisation ist eine der wichtigsten Aufgaben unseres Hörsystems. Die Position von tieffrequenten Schallquellen wird vor allem auf der Basis von interauralen Zeitdifferenzen (ITD) bestimmt. Die Verarbeitung solcher ITDs findet in der medialen oberen Olive (MSO), einer Struktur des auditorischen Hirnstamms statt (Goldberg and Brown, 1969; Yin and Chan, 1990; Spitzer and Semple, 1995), die zum ersten Mal in der aufsteigenden Hörbahn binaurale akustische Information verarbeitet. Die Zellen in der MSO bekommen von beiden Ohren erregende und hemmende Eingänge. Ein zeitlich präzise abgestimmtes Zusammenspiel dieser vier Eingänge sorgt für die richtige Einstellung der ITD-Empfindlichkeit in der Wüstenrennmaus (Brand et al., 2002). Die Koinzidenz der erregenden Eingängen alleine erzeugt eine ITD-Sensitivität, die bei ca. 0 ITD ihre maximale Antwort hat. Dadurch liegt die maximale Steigung der ITD-Funktion außerhalb des physiologisch relevanten Bereiches. Die Inhibition sorgt dafür, dass die maximale Antwort in den contralateralen Bereich verschoben und somit die maximale Steigung der ITD-Funktion auf den Bereich der physiologisch relevanten ITDs abgestimmt wird. Die glyzinergen inhibitorischen Projektionen zur MSO der Wüstenrennmaus sind vor Hörbeginn noch diffus verteilt und innervieren Somata und Dendriten gleichermaßen. Weniger als zwei Wochen nach Hörbeginn sind diese hemmenden Eingange jedoch auf die Somata der MSO-Neurone beschränkt (Seidl, 1999). Diese Beschränkung ist abhängig von binauraler Aktivität (Kapfer, 1999). In der vorliegenden Arbeit wird gezeigt, dass diese Eliminierung der dendritischen inhibitorischen Eingänge in der Wüstenrennmaus durch die Aufzucht in omnidirektionalem weißem Rauschen während einer kritischen Periode nach Hörbeginn unterdrückt werden kann. Für die normale Entwicklung der räumlichen Verteilung der glyzinergen Synapsen in der MSO ist also normale akustische Erfahrung notwendig. Bei Tieren, die ITDs nicht zur Schalllokalisation verwenden, kommt es zu keiner solchen Entwicklung. Vor Hörbeginn und auch im Erwachsenenstadium sind die inhibitorischen Eingänge auf den Zellen der MSO gleichmäßig über Soma und Dendriten verteilt. Als weiteres Ergebnis wird beschrieben, dass es eine Veränderung der ITD-Empfindlichkeit nach Hörbeginn gibt. Die Abstimmung der maximalen Steigungen der ITD-Funktionen auf den physiologischen Bereich nach Hörbeginn unterbleibt, wenn die räumlichen akustischen Signale durch weißes Rauschen während der kritischen Periode maskiert werden. Diese Entwicklung korreliert mit der Verteilung der glyzinergen Synapsen an MSO-Neuronen. Werden erwachsene Tiere weißem Rauschen ausgesetzt, so kommt es zu einer Änderung der ITD-Empfindlichkeit, die reversibel ist, aber nicht mit der unterdrückten Entwicklung nach Hörbeginn vergleichbar ist. Diese Arbeit zeigt, dass die korrekte strukturelle Entwicklung inhibitorischer Synapsen notwendig ist um die biophysikalische Grundlage für Schalllokalisationsmechanismen zu schaffen. Diese Entwicklung ist abhängig von der Erfahrung räumlicher akustischer Signale. Somit ist sie ein Beispiel für ein System, das sich direkt durch die Information die es später verarbeitet, selbst abstimmt und optimiert.