O complexo Mre11 (Mre11/Rad50/Nbs1) é uma componente chave da resposta celular ao dano ao DNA em humanos e recentes observações sugerem que estas proteínas são em parte responsáveis pela interface ente o ensoreamento do dano ao DNA, seu reparo e as funções das proteínas envolvidas nos pontos de checagem do ciclo celular. Em Aspergillus nidulans, a partir de um screening para o isolamento de letais sintéticos na ausência de dineína, o gene sldIRAD50 foi clonado como um desses letais sintéticos através da complementação do fenótipo de deficiência de conidiação do mutante. Foi identificada uma transversão G-C na posição 2509 (Ala-692-Pro) no mutante sldI1444 a qual está presente na região de dobradiça da proteína. Essa mutação causa sensibilidade a vários agentes mutagênicos. Uma linhagem mutante sldIRAD50::pyrG foi construída a qual apresentou também vários defeitos na reposta celular ao dano ao DNA incluindo sensibilidade a várias drogas mutagênicas, defeito no ponto de checagem de replicação do DNA e na viabilidade dos ascosporos. Além disso, o gene sldIRAD50 interage geneticamente com bimEAPC1 para o controle do spindle pole checkpoint durante a segregação cromossômica sugerindo um novo papel para o complexo Mre11. Em atuação paralela com o complexo Mre11, duas proteínas quinases ditas apicais, ATM e ATR coordenam a transdução do sinal do dano ao DNA para proteínas efetoras do reparo. A proteína ATM está mutada na síndrome de instabilidade cromossômica herdada Ataxia Telangiectasia. Para a caracterização do homologo de ATM em A. nidulans AtmA, uma linhagem mutante atmAATM foi isolada. Esse mutante apresentou falha na reposta ao dano ao DNA, como seus homólogos em vários outros organismos mostrando defeitos no ponto de checagem intrafase S e G2/M, além de sensibilidade a camptothecin e bleomicina. Ainda, o extrato protéico bruto desse mutante não foi capaz de fosforilar o homologo de NBS1 em A. nidulans, ScaA. Além das conhecidas funções de ATM na resposta ao dano ao DNA, foi verificado que o mutante atmAATM apresentou uma acelerada cinética de divisão nuclear e severos defeitos no estabelecimento e manutenção do eixo de crescimento polarizado, evidenciando uma função ainda não descrita para ATM no crescimento polar. Provavelmente, AtmA regula a função e/ou localização de proteínas chaves para a formação do eixo de polarização. Diante disso, para investigar as vias metabólicas que são controladas por esse gene, o perfil transcricional do mutante atmAATM, em comparação com a linhagem selvagem foi verificado em diferentes condições de crescimento. Os resultados indicaram um importante papel da via das pentoses fosfato na proliferação celular monitorada pela AtmA. Além disso, foram identificados vários genes com a expressão do mRNA diminuída envolvidos no crescimento polarizado, na síntese de ácido fosfatídico e de ergosterol e no tráfico intracelular, secreção e transporte vesicular. Buscando identificar genes que participam da resposta celular ao dano ao DNA causado pela droga anti topoisomerase I, camptothecin, foram utilizados filtros de macroarray de A. nidulans contendo 2787 genes deste organismo para monitorar a expressão gênica da linhagem selvagem e do mutante uvsBATR, num experimento de indução com CPT por 30, 60 e 120 minutos. Os resultados revelaram um total de 1512 e 1700 genes modulados na linhagem selvagem e uvsBATR respectivamente, em pelo menos um ponto experimental. Seis desses genes que apresentaram aumento da expressão de mRNA na linhagem selvagem e diminuição da linhagem uvsBATR foram caracterizados: fhdA (que codifica para uma proteína com domínio fork-head associated), tprA (uma proteína hipotética que apresenta o domínio tetratrico peptide repeat), mshA (um homólogo MutS6 envolvido em mismatch repair), phbA (um homólogo da prohibitina), uvsCRAD51 e cshA (homólogo da proteína CSB envolvida no reparo por excisão de nucleotídeos e ligada a Síndrome de Cockayne). A indução transcricional desses genes na presença de CPT requer a função de uvsBATR. Estes genes foram deletados e surpreendentemente apenas uvsCRAD51 apresentou sensibilidade a CPT, enquanto os outros mostraram sensibilidade a outros agentes que causam dano ao DNA e estresse oxidativo. Além disso, com exceção de uvsCRAD51, a deleção desses genes leva a supressão parcial da sensibilidade a menadiona e paraquat do mutante uvsBATR. Esses resultados indicaram um comportamento heterogêneo de sensibilidade durante o crescimento na presença de agentes que causam dano direto ou indireto ao DNA, evidenciando que o perfil transcricional não é determinante para predizer a função de um gene na proteção da célula a determinada droga que causa dano ao DNA. / The Mre11 protein complex (Mre11/Rad50/Nbs1) has emerged as a central component in the human cellular DNA damage response, and recent observations suggest that these proteins are at least partially responsible for the linking of DNA damage detection to DNA repair and cell cycle checkpoint functions. In Aspergillus nidulans, the sldI1444D mutant was isolated in a screen for dynein synthetic lethals. The sldIRAD50 gene was cloned by complementation of the sporulation deficiency phenotype of this mutant. A transversion G-C at the position 2509 (Ala-692-Proamino acid change) in the sldI1444D mutant causes sensitivity to several DNAdamaging agents. The mutation sldI1 occurs at the CXXC hinge domain of Rad50. An inactivation strain sldIRAD50::pyrG was constructed. Besides sensitivity to a number of DNA-damaging agents, this deletion strain was also impaired in the DNA replication checkpoint response and in ascospore viability. Also, sldIRAD50::pyrG geneticaly interacted with bimEAPC1, acting in the spindle pole checkpoint control during segregation, suggesting a new possible role of Mre11 complex. In parallel to the Mre11 complex, two apical quinases ATM and ATR respond to DNA damage and transduce the signal to effector proteins. In humans, mutations in ATM cause the devastating neurodegenerative disease Ataxia Telangiectasia. Here we characterized the homolog of ATM (AtmA) in the filamentous fungus A. nidulans. The deletion strain atmA presented defects in the DNA damage response as previously shown in other model organisms including intra S-phase and G2/M checkpoint defects, sensitivity to camptothecin and bleomycin. Also, the crude extract from the mutant strain did not phosphorylate the NBS1 homologue ScaA. In addition to its expected role in the DNA damage response, the atmA mutant showed increased nuclear division kinetics and severe defects in polarized hyphal growth, indicating a novel feature for the ATM gene. Probably, AtmA regulates the function and/or localization of landmark proteins required for the formation of a polarity axis. We extended these studies by investigating which pathways are controlled by AtmA during proliferation and polar growth by comparatively determining the transcriptional profile of A. nidulans wild type and atmA mutant strains in different growth conditions. Our results indicated an important role of the pentose phosphate pathway in the fungal proliferation during endogenous DNA damage and polar growth monitored by the AtmA kinase. Furthermore, we identified several genes that have decreased mRNA expression in the atmA mutant that are involved in the formation of polarized hyphae and control of polar growth; in the biosynthesis of phosphatidic acid and ergosterol; and intracellular trafficking, secretion, and vesicular transport. In order to identify genes that responded to the DNA damage mediated by the anti- toposomerase I drug, camptothecin, we used an A. nidulans macroarray carrying sequences of 2,787 genes from this fungus to monitor gene expression of both wild-type and uvsBATR in a time-point experiment where mycelium was exposed to 60, 90 and 120 minutes to the drug. The results revealed a total of 1,512 and 1,700 genes in the wild-type and uvsBATR deletion mutant strain that displayed statistically significant difference in at least one experimental time-point. We characterized six genes that have increased mRNA expression in the presence of CPT in the wild-type strain relative to the uvsBATR mutant strain: fhdA (encoding a fork head associated domain protein), tprA (encoding a hypothetical protein that contains a tetratrico peptide repeat), mshA (encoding a MutS homologue involved in mismatch repair), phbA (encoding a prohibitin homologue), uvsCRAD51 (the homologue of the RAD51 gene), and cshA (encoding a homologue of the excision repair protein ERCC-6 [Cockaynes syndrome protein]). The induced transcript levels of these genes in the presence of CPT required uvsBATR. These genes were deleted, and surprisingly, only the uvsCRAD51 mutant strain was sensitive to CPT; however, the others displayed sensitivity to a range of DNA-damaging and oxidative stress agents. Moreover, with the exception of UvsC, deletion of each of these genes partially suppressed the sensitivity of the uvsB strain to menadione and paraquat. These results indicated a very complex and heterogeneous sensitivity behavior during growth in the presence of agents that directly or indirectly cause DNA damage and the transcriptional response to DNAdamaging agents does not necessarily identify the genes that protect against these agents.

Glicose e oxigênio são moléculas essenciais para a maioria dos organismos vivos. Além de sua importância nos processos de produção de energia - glicose como fonte de carbono e energia e oxigênio como aceptor dos elétrons doados por NADH e FADH2 - estes dois compostos funcionam como efetuadores, modulando vários processos metabólicos e fisiológicos nas células. Visto que a mitocôndria é um dos alvos afetados pelas disponibilidades destas duas moléculas, nós isolamos e seqüenciamos o genoma mitocondrial de Trichoderma reesei, um fungo multicelular empregado neste trabalho como sistema modelo. Foi estudado o efeito da variação de concentração de glicose e oxigênio sobre a expressão de transcritos do genoma mitocondrial, bem como sua implicação no metabolismo de glicose. São apresentadas análises da expressão gênica de aproximadamente 2000 transcritos de T. reesei submetido a concentrações limitantes de oxigênio dissolvido, realizadas com o emprego da técnica de microarrays de cDNA. Pelo menos 330 transcritos foram diferencialmente expressos em função da disponibilidade de oxigênio. Aqueles envolvidos nos processos de síntese protéica e divisão celular foram regulados negativamente, enquanto transcritos relacionados com funções de defesa celular e síntese de RNA foram positivamente regulados. Uma fração substancial de outros genes afetados pela baixa disponibilidade de oxigênio não possui, atualmente, funções celulares conhecidas. Esta observação deve contribuir para a posterior anotação funcional do genoma de T. reesei. Também foram identificados reguladores transcricionais diferencialmente expressos em baixas tensões de oxigênio. O perfil de expressão destes reguladores aponta-os como potenciais candidatos ao envolvimento com a expressão de genes afetados pela disponibilidade de oxigênio. / Glucose and oxygen are essential molecules in most of living organisms. In addition to their importance in production of energy - glucose as a carbon and energy source and oxygen as an acceptor of electrons donated by NADH and FADH2 - both molecules function as effectors modulating various metabolic and physiological processes in the cell. Because one of the targets affected by both molecules is the mitochondrion, we isolated and sequenced the mitochondrial genome of Trichoderma reesei, a multicellular fungus that is used in this study as a model system. The effect of varying the concentration of glucose and oxygen on the expression of the transcripts of the mitochondrial genome, and its implication on the metabolism of glucose, was studied. Gene-wide expression analyses of nearly 2000 transcripts of T. reesei under limited concentration of dissolved oxygen, using cDNA microarry technique, are presented. At least 330 transcripts were differentially expressed with respect to oxygen availability. Those involved in protein synthesis and cell division processes were downregulated, while transcripts involved in cell defense and RNA synthesis were upregulated. A substantive fraction of other anaerobically affected genes have currently unknown cellular roles, and these results should therefore contribute to further functional annotation of the genome. ln addition, we have identified transcriptional regulators that are differentially expressed at a low oxygen tensions. The expression profile of these regulators points them out as potential candidates involved in the expression of genes affected by oxygen availability.

Pancreatic ductal adenocarcinoma (PDAC) has an extremely poor prognosis. To improve the prognosis, novel molecular markers and targets for earlier diagnosis and adjuvant and/or neoadjuvant treatment are needed. Recent advances in human genome research and high-throughput molecular technologies make it possible to cope with the molecular complexity of malignant tumors. With DNA array technology, mRNA expression levels of thousand of genes can be measured simultaneously in a single assay. As several studies using microarrays in PDAC have already been published, this review attempts to compare the published data and therefore to validate the results. In addition, the applied techniques are discussed in the context of pancreatic malignancies. / Dieser Beitrag ist mit Zustimmung des Rechteinhabers aufgrund einer (DFG-geförderten) Allianz- bzw. Nationallizenz frei zugänglich.

In the European Union an environmental risk assessment is required for the registration of new chemicals, biocides, pesticides and pharmaceuticals. In order to avoid the release of potential hazardous substances, various ecotoxicity tests are performed, including acute and chronic fish tests. As a consequence of the new program of the European Union “Registration, Evaluation and Authorisation of Chemicals” (REACH) the number of animal experiments for environmental risk assessment is expected to increase remarkably within the next years. On the other hand there is a strong societal demand for reducing the number of animal tests by using alternative in vitro models. According to EU directives, investigations using non-human vertebrate embryos are considered pain free in vitro methods and are therefore accepted as alternatives to animal experiments. For the acute fish test, the Danio rerio embryo test (DarT) has been established as a replacement method and included in national regulations at least for waste water (German Waste Water Dues Law). However, no alternatives for chronic fish tests are currently available. The overall goal of this thesis was to work towards such a replacement by extending DarT zu Gene-DarT. Toxicants will initially interact at the molecular level with consequences for physiology, fitness and survival. The analysis of gene expression patterns may unravel elements of these molecular events before any phenotypic changes are visible. The hypothesis of this thesis therefore was that chemical-sensitive genes in embryos exposed in a conventional DarT may indicate toxic impact of substances at sub-acute concentrations and thus enhance the sensitivity of the embryo toxicity test. Furthermore, unlike the conventional DarT-endpoints, gene expression analysis will provide insights into mechanistic processes underlying toxicity. The 3,4-dichloroaniline (3,4-DCA), which is used as a reference compound in the DarT, was selected as model chemical in this thesis. In a first step, differentially expressed genes in embryos exposed to 3,4-DCA were identified by microarray technology and RT-PCR techniques. Six dose-dependent significant differentially expressed genes were identified. These genes were involved in biotransformation pathways (cyp1a, ahr2), stress response (nrf2, maft, ho-1) and cell cycle control (fzr1). Differential expression upon 3,4-DCA exposure was detected below the LOEC (lowest observed effect concentration = 6.2 µM) of survival or developmental disorders of the embryo test (0.78 µM and above). For the validation of stage specific sensitivity, genes were also analysed in post-hatched stages. Extension of exposure to post-hatched stages resulted in a differential expression at lower concentrations as for the embryonic stages, indicating an improved sensitivity due to stage-specific sensitivity or exposure time. To confirm the adaptive function of the 3,4-DCA-sensitive genes, embryonic mRNA abundance was experimentally manipulated by knock down and overexpression. By injection of sense (mRNA) or antisense (siRNA) RNA in one-cell-stages of embryos, the transcript levels of genes were transiently enhanced or repressed in embryos exposed to 3,4-DCA. mRNA injection of the genes cyp1a, ho-1 and nrf2 reduced the number of embryos with 3,4-DCA-induced malformations. In contrast, siRNA injections for the same genes led to an increase in the severity and frequency of developmental disorders. The results clearly indicate the adaptive functions of the investigated genes or their corresponding proteins. This study demonstrates that the analysis of chemical-sensitive gene expression shows the potential to increase the sensitivity of conventional toxicity tests. The analysis of gene expression also provides additional mechanistic information for toxic action, e.g. in the presented study, the involvement of Ah-receptor regulated pathways as an adaptive response. Furthermore, the presented data indicate that functional manipulations, using mRNA and siRNA-injection, are suitable to evaluate the role of differentially expressed genes for toxicity.

Staphylococcus aureus ist einer der bedeutendsten Erreger boviner Mastitiden. Die Vielgestaltigkeit der Resistenzmuster und Virulenzfaktoren seiner Stämme macht ihn zu einem Problemkeim aus therapeutischer und prophylaktischer Sicht. Seine Fähigkeit zur Bildung hitzestabiler Enterotoxine verleiht ihm lebensmittelhygienische Relevanz. Mehrfachresistente Stämme stellen gefährliche Zoonose-Erreger dar. Ziel der durchgeführten Untersuchung war es daher, Aufschluss über Resistenzdeterminanten und Virulenzfaktoren der in Thüringer Milchviehherden vorkommenden Staphylococcus aureus zu erhalten, wobei eine Microarray-gestützte Genotypisierung zum Einsatz kam. Weiterhin sollte analysiert werden, ob der Genotyp der Isolate mit dem Phänotyp korreliert. In 34 Thüringer Milchviehherden wurde der gesamte Bestand der laktierenden Kühe zweimal auf Basis von Viertelgemelksproben bakteriologisch untersucht. Die Beurteilung der Kulturen erfolgte im Nativausstrich nach 48-stündiger Bebrütung und zusätzlich nach Voranreicherung in einer Glucose-Bouillon mit anschließender 24-stündiger Bebrütung. Staphylococcus aureus-positiv waren 1902 von insgesamt 81 567 Milchproben. Aus diesen wurden 189 für die Herden repräsentative Isolate ausgewählt und mittels Microarray-Technologie umfassend charakterisiert und klassifiziert. Zudem wurde der Phänotyp der Isolate auf Äskulin- und Columbia-Blutagar erfasst und das Resistenzverhalten mittels Agardiffusionstest ermittelt. Die 189 typisierten Staphylococcus aureus konnten elf verschiedenen klonalen Komplexen (CC) zugeordnet werden. Der Großteil der Isolate (80,4 %) zählte zu CC133, CC151 und CC479. Diese Isolate besaßen mit einer Ausnahme das Leukozidin-Gen lukF-P83/lukM. Die übrigen Isolate, die negativ auf lukF-P83/lukM getestet wurden, gehörten acht vergleichsweise sporadisch vorkommenden CC (CC7, CC9, CC20, CC45, CC50, CC97, CC101, CC398) an. In nur 0,7 % der zu den drei dominanten CC zählenden Isolate war das Beta-Laktamase-Gen blaZ vorhanden, während es bei 54,1 % der sporadisch vorkommenden CC detektiert wurde. Das Methicillin-Resistenzgen mecA wurde bei lediglich vier Isolaten (2,1 %) nachgewiesen, die alle CC398 angehörten. Sie verfügten neben Resistenzen gegenüber β-Laktam-Antibiotika über eine Tetrazyklin-Resistenz. Darüber hinaus wurde in einem Isolat das Makrolid/Lincosamid/Streptogramin-Resistenz vermittelnde vgaA und in einem Isolat das Aminoglykosid-Resistenz vermittelnde aacA-aphD detektiert. Humanmedizinisch relevante Enterotoxin-, Exfoliatin- oder PVL-Gene wurden in den vier Methicillin-resistenten Staphylococcus aureus (MRSA) nicht gefunden. Im Agardiffusiontest zeigten diese Isolate eine Penicillin- und eine Tetrazyklin-Resistenz, jedoch keine Resistenz gegenüber Oxacillin, welches als MRSA-Marker gilt. Die Gene der klassischen, humanmedizinisch bedeutsamen Enterotoxine A, B und C waren bei 12,7 % der Isolate vorhanden, wohingegen die Gene von Enterotoxin D und E nicht vorkamen. Insgesamt fanden sich Enterotoxin-Gene bei 78,3 % der typisierten Staphylococcus aureus, wobei die für Enterotoxin G, I, M, N, O und U kodierenden dominierten. Phänotypisch unterschieden sich die CC bezüglich Hämolyse und Pigmentierung, wobei alle CC398-Isolate als eierschalenfarben mit doppelzoniger Hämolyse auftraten. Hämolysin-Gene besaßen alle Isolate, ein Zusammenhang zu den phänotypisch ausgeprägten Hämolysezonen bestand jedoch nicht. Die vorliegende Untersuchung zeigt, dass in Thüringer Milchviehbeständen zwei epidemiologisch unterschiedliche Varianten von Staphylococcus aureus existieren. Die in dieser Studie dominierenden, lukF-P83/lukM-positiven CC133, CC151 und CC479 verursachten einen Großteil der Infektionen und gelten als auf das Euter beschränkte Erreger. Sie können daher als „euterassoziiert“ angesehen werden. Dagegen verfügten die anderen in dieser Untersuchung detektierten, lukF-P83/lukM-negativen CC über Charakteristika „umweltassoziierter“ Keime. Sie besitzen ein breites Wirtsspektrum und treten auch außerhalb des bovinen Euters in der Umgebung der Kühe auf. Die Prüfung auf lukF-P83/lukM erwies sich als zuverlässige Methode, zwischen beiden epidemiologischen Varianten zu unterscheiden. Folglich lässt die An- oder Abwesenheit dieser Genkombination einen Rückschluss auf die in der Herde verbreiteten CC zu. Das ermöglicht die Berücksichtigung der CC-spezifischen Erreger-Eigenschaften bei der Etablierung von Sanierungsprogrammen, die somit effizient gestaltet werden können. MRSA waren in Thüringer Milchviehbeständen wenig verbreitet und nur schwach mit Resistenzdeterminanten und humanmedizinisch bedeutsamen Pathogenitätsfaktoren ausgestattet. Diese MRSA aus Rohmilchproben sind daher nicht mit multiresistenten Isolaten aus der Humanmedizin zu vergleichen. Gene für humanmedizinisch relevante Enterotoxine, für die ein Zusammenhang mit Lebensmittelintoxikationen belegt ist, wurden selten, andere Enterotoxin-Gene jedoch häufig nachgewiesen.

Background: Recently high-throughput sequencing (HTS) using next generation sequencing techniques became useful in digital gene expression profiling. Our study introduces analysis options for HTS data based on mapping to miRBase or counting and grouping of identical sequence reads. Those approaches allow a hypothesis free detection of miRNA differential expression. Methods: We compare our results to microarray and qPCR data from one set of RNA samples. We use Illumina platforms for microarray analysis and miRNA sequencing of 20 samples from benign follicular thyroid adenoma and malignant follicular thyroid carcinoma. Furthermore, we use three strategies for HTS data analysis to evaluate miRNA biomarkers for malignant versus benign follicular thyroid tumors. Results: High correlation of qPCR and HTS data was observed for the proposed analysis methods. However, qPCR is limited in the differential detection of miRNA isoforms. Moreover, we illustrate a much broader dynamic range of HTS compared to microarrays for small RNA studies. Finally, our data confirm hsa-miR-197-3p, hsa-miR-221-3p, hsa-miR-222-3p and both hsa-miR-144-3p and hsa-miR-144-5p as potential follicular thyroid cancer biomarkers. Conclusions: Compared to microarrays HTS provides a global profile of miRNA expression with higher specificity and in more detail. Summarizing of HTS reads as isoform groups (analysis pipeline B) or according to functional criteria (seed analysis pipeline C), which better correlates to results of qPCR are promising new options for HTS analysis. Finally, data opens future miRNA research perspectives for HTS and indicates that qPCR might be limited in validating HTS data in detail.:Background; Methods; Results; Discussion; Conclusions

Streptococcus (S.) agalactiae sind grampositive, in Ketten gelagerte Kokken, die auf Blutagar eine Hämolyse zeigen. Aufgrund ihrer Zugehörigkeit zur Lancefield-Gruppe B werden sie auch als Gruppe B Streptokokken (GBS) bezeichnet (Hof, 2005). GBS sind die Hauptursache von Sepsis, Meningitis und Pneumonie bei Neugeborenen (Schrag, et al., 2000). Die Arbeit beschäftigte sich mit der Genotypisierung von GBS. Darüber hinaus konnten auch Einblicke in die Phylogenese sowie die Populationsstruktur von GBS gewonnen werden. Ziel war es, einen DNA-Microarray zu entwickeln und zur Genotypisierung von GBS einzusetzen. Während der Evaluierung des DNA Microarray konnten stammspezifische Muster beobachtet werden, diese wurden durch bereits etablierte Typisierungmethoden (MLST) bekannten Genotypen zugeordnet. Die Ergebnisse wurden in einer Datenbank zusammengefasst. Mithilfe der Datenbank konnte die Software zur Auswertung entwickelt werden. (siehe http://alere-technologies.com/en/products/lab-solutions.html). Der DNA-Microarray trägt Sonden für GBS-spezifische Virulenzfaktoren und Oberflächenmarker. Für die auf dem Microarray basierende Typisierung wurden 11 über das ganze Genom verteilte Gene bzw. Gencluster (bac, alp, pil1 locus, pepS8, fBsB, capsule locus, hylB, abiG-I/-II plus Q8DZ34, pil2 locus, nss plus srr plus rogB2 und rgfC/A/D/B) ausgewählt. Ubiquitär vorkommende, konservierte Gene (z. B. cylD/cylE) eignen sich nicht als Marker für eine Typisierung, wurden aber als Kontrollen und zur Normierung eingeschlossen. Zur vollständigen Charakterisierung wurden außerdem Sonden für hochmobile plasmidgebundene Resistenzgene wie z. B. erm(A), erm(B), erm(C), tet(M), emrB/qacA aufgetragen (Aracil, et al., 2002; Betriu, et al., 2003; Uh, et al., 2001). Diese Gene sind nicht für GBS spezifisch. Sie eignen sich z. B. für eine Unterscheidung einzelner Isolate, nicht jedoch für die Unterteilung der GBS Population in verschiedene Stämme. Für die vorliegende Arbeit wurden insgesamt 448 klinische Isolate von GBS ausgewählt und untersucht. Darunter waren Isolate, die schwerwiegende Erkrankungen wie Sepsis und Meningitis verursacht haben, Isolate aus lokalen Infektionen sowie Isolate von asymptomatischen/gesunden Trägern. Zu Vergleichszwecken wurden außerdem Isolate aus der Veterinärmedizin (von bovinen Mastitisfällen) und humane Isolate aus einer geographisch weit entfernten Region (Trinidad und Tobago) genotypisiert. Für 36 ausgewählte Isolate mit repräsentativen Hybridisierungsmustern wurde zusätzlich eine Typisierung mittels Multilocus-Sequenztypisierung (MLST) (Jones, et al., 2003) durchgeführt. Die Hybridisierungsmuster vom Microarray wurden mit Daten aus diesem bereits etablierten Typisierungssystem verbunden. Durch die Verknüpfung beider Methoden konnte eine Einteilung der GBS Isolate in verschiedene Stämme vorgenommen werden. Mit Hilfe des eBURST-Algorithmus wurde gezeigt, dass einige Hybridisierungsmuster sich zu Gruppen zusammenfügen. Dieses Verfahren veranschaulicht die Populationsstruktur und beschreibt die genetische Vielfalt. Mit den 11 definierten Markern konnten die untersuchten Isolate 76 verschiedenen Stämmen bzw. „hybridization profiles“ (HP) zugeordnet werden. Die Einteilung beruht auf dem Fehlen bzw. Vorhandensein einzelner Gene/Gencluster bzw. deren allelischen Varianten. Diese Stämme korrelieren mit den durch MLST definierten klonalen Komplexen (CC). Isolate mit identischen oder ähnlichen Hybridisierungsprofilen gehören zum selben CC. Dagegen können Isolate mit einem ähnlichen MLST-Profil verschiedene Hybridisierungsmuster zeigen. Es konnte außerdem häufig beobachtet werden, dass ansonsten ähnliche Stämme sich in einzelnen Merkmalen, z. B. Kapsel-Genen, alp- oder pili-Genen, voneinander unterscheiden, und dass diese Gene unabhängig voneinander variieren. Zusätzlich zeigten einige ubiquitäre Gene/Gencluster, die sich in den publizierten Genomsequenzen immer an derselben Position befinden, zahlreiche verschiedene Allele. Welches Allel in einem gegebenen Stamm gerade vorliegt, scheint dabei eher zufällig zu sein. Eine Erklärung dieses Phänomens könnte in vergangenen Rekombinationsereignissen liegen. Auch eine konvergente Evolution könnte diskutiert werden. Ähnliche Stämme/ „hybridization profiles“ wurden in Analogie zu den MLST-definierten klonalen Komplexen zu Gruppen zusammengefügt. Das bedeutet jedoch nicht notwendigerweise eine direkte Verwandtschaft der Isolate im Sinne des Vorhandenseins eines unmittelbaren gemeinsamen Vorfahren. Die Typisierung sowohl über den DNA-Microarray als auch über die MLST kann nicht die „wahre“ Phylogenese im Rahmen der Evolutionsgeschichte und Herkunft widerspiegeln. Sie stellt lediglich ein zufälliges Modell dar, ein Ordnungssystem im Sinne eines genetischen Fingerabdrucks, das einen Vergleich von Isolaten, aber keine Rückschlüsse über deren Abstammung und Herkunft erlaubt.Die untersuchten GBS Isolate konnten in fünf klonale Komplexe (CC19, CC23, CC26, CC103, CC130) eingeteilt werden, deren Häufigkeit unterschiedlich war. Deutsche humanmedizinische Isolate konnten vorwiegend CC19 zugeordnet werden. Karibische humanmedizinische Isolate sind zumeist CC19 und CC23 zugehörig. Bovine Isolate gehören meist zu CC19 und CC103. Unter humanen Isolaten ist CC103 rar. Vermutlich basierend auf der geografischen und wirtsspezifischen Herkunft der untersuchten GBS Isolate gibt es Unterschiede in der Populationsstruktur. In der vorliegenden Arbeit war CC19 der am häufigsten gefundene und außerdem ein genetisch besonders inhomogener CC. Er besteht aus mehreren unterschiedlichen, bisher als eigenständig angesehenen CCs (darunter CC1, CC17, CC19 und CC22). Diese werden von dem zur MLST-Verwandtschaftsanalysen verwendeten eBURST-Algorithmus zu CC19 zusammengefasst, seit die MLST-Profile von 'missing links' zwischen den CCs identifiziert wurden, da eBURST „gemeinsame Vorfahren“ nicht von durch horizontalem Gentransfer bzw. durch Hybridisierungen entstandenen „Chimären“ unterscheiden kann. Da diese Komplexe klar unterscheidbare Hybridisierungsmuster aufweisen, wurden sie hier als CC19/01, CC19/17, CC19/19 und CC19/22 bezeichnet. Einzelne Gene traten in Gruppen von Isolaten aus verschiedener Herkunft unterschiedlich häufig auf. So fand sich der Virulenzfaktor scpB in 412 von 418 humanen Isolaten (98,6 %), aber nur in 10 von 21 Rinderisolaten (48 %). Ferner ließ sich beobachten, dass invasive Isolate weniger wahrscheinlich abiGI-/II und Q8DZ34 tragen, jedoch häufiger pil1 locus, fbsB (515) und Kapseltyp III sowie pil2b, nss/srr und rgf (COH1 like) aufweisen. Einige dieser Marker erscheinen zusammen in CC19/17-Stämmen, welche häufig bei invasiven Krankheitsverläufen beobachtet werden. CC19 (incl. ST01, ST17, ST19) konnte bei neonatalen Sepsis-Fällen in verschiedenen geografischen Regionen isoliert werden (Brzychczy-Wloch, et al., 2012; Ryu, et al., 2014; Sorensen, et al., 2010; Strakova, et al., 2010; Tien, et al., 2011). Zusätzlich wurden andere Virulenzfaktoren wie speM (Exotoxin M) und das cyl-Operon (beta-Hämolysin) untersucht. In lediglich sieben Isolaten wurde speM nachgewiesen. Das cyl-Operon konnte in allen Isolaten gefunden werden, sein Nachweis ist daher für eine Vorhersage der Virulenz eines gegebenen Isolates nicht hilfreich. Es konnte kein einzelner Faktor zur definitiven Unterscheidung zwischen invasiven Isolaten und Trägerisolaten bestimmt werden. Für die Virulenz eines Isolates ist wahrscheinlich nicht das bloße Vorhandensein oder Fehlen eines bestimmten Genes ausschlaggebend, sondern dessen Expression in vivo. Wichtig wäre in diesem Zusammenhang auch die genaue Betrachtung der Sequenz eines als Virulenzfaktor angesehenen Genes sowie die Untersuchung der zugehörigen regulatorischen Gene. Über den Nachweis der Gene erm(A), erm(B) und erm(C) konnte eine Aussage über die Macrolid-/Clindamycinresistenz eines GBS Isolates getroffen werden. Bei keinem der karibischen Isolate wurden erm Gene nachgewiesen. Innerhalb der deutschen GBS Population wurde erm(B) am häufigsten beobachtet. Die Gene erm(A) und erm(C) waren in humanen Isolaten selten und wurden in bovinen Isolaten überhaupt nicht gefunden. Das Tetracyclinresistenzgen tet(M) wurde häufig in humanen Isolaten und sehr selten in veterinärmedizinischen Isolaten gefunden. Für weiterführende Untersuchungen könnte die beschriebene Typisierungsmethode verfeinert werden. So lassen sich z. B. die oben beschriebenen 11 ausgewählten Typisierungsmarker des Microarrays mit denen der MLST zu einem 18 Marker-System verknüpfen. Daneben können auch erm-, cad-, mer- oder tet-Gene zur Feststellung oder zum Ausschluss der Identität verwandter Isolate in vitro oder in silico verwendet werden. Mit der nun einsatzbereiten Genotypisierungsmethode können in Zukunft weitere Studien zur Untersuchung regionaler und wirtsspezifischer Unterschiede der GBS Population durchgeführt werden. In dieser Arbeit konnte gezeigt werden, dass der DNA-Microarray stabile und reproduzierbare Resultate erbringt. Es kann ein detaillierter Befund erstellt werden, die Ergebnisse sind mit denen anderer Typisierungsmethoden und der Genomsequenzierung vergleichbar. Jedoch steht mit dem DNA-Microarray ein wesentlich unkomplizierteres und schnelleres Procedere zur Verfügung, welches zudem geringere Kosten verursacht.:1. Einleitung 5 1.1 Gegenstand der Untersuchung 5 1.2 Entdeckungsgeschichte 6 1.3 Klinische Bedeutung 7 1.4 Veterinärmedizinische Bedeutung 9 1.5 Stand der Forschung mit Hinblick auf Typisierung von GBS 9 1.6 Ziele der vorliegenden Untersuchung 10 1.7 Vorgehensweise 11 1.7.1 Untersuchungsmaterial 11 1.7.2 Untersuchungsmethode 11 1.8 Zur Typisierung ausgewählte Marker 12 2. DNA Microarray-Based Typing of Streptococcus agalactiae Isolates 15 2.1 Abstract 16 2.2 Introduction 17 2.3 Materials and methods 18 2.3.1 Bacterial isolates 18 2.3.2 Ethics statement 18 2.3.3 Preparation of genomic DNA 19 2.3.4 MLST 19 2.3.5 Microarray design and protocol optimization 19 2.3.6 Microarray procedures 20 2.3.7 eBURST 21 2.4 Results 22 2.4.1 Typing GBS by MLST 22 2.4.2 Genotyping GBS by microarray hybridization 22 2.4.3 Population structure 25 2.4.4 Detection of antibiotic resistance markers 25 2.4.5 Detection of heavy metal resistance markers 26 2.5 Discussion 27 2.6 Acknowledgments 30 2.7 References 31 2.8 Tables and figures 33 3. Zusammenfassung 45 3.1 Zusammenfassung 45 3.2 Summary 49 4. Korrespondenz mit dem Editor 53 4.1 Hinweise des Editors und der Gutachter 53 4.2 Antworten an den Editor 56 4.3 Endgültige Annahme 58 Anhang 61 / Streptococcus (S.) agalactiae are Gram-positive, chain-forming cocci, which show hemolysis on blood agar. They are also referred to group B streptococci (GBS) because of their affiliation to Lancefield-group B (Hof, 2005). GBS are the main cause of sepsis, meningitis and pneumonia in newborns (Schrag, et al., 2000). The work focused on genotyping of GBS. The aim was to develop a DNA microarray and to use it for epidemiological typing of GBS. During the evaluation of the microarray, strain-specific patterns could be observed and these patterns assigned to genotypes as defined by other typing methods (MLST). The results were summarized in a database that subsequently was developed into software for automated analysis of experiments (http://alere-technologies.com/en/products/lab-solutions.html). The DNA microarray carries probes for GBS specific virulence factors and surface markers. For the microarray-based typing, 11 genes or gene clusters were selected that are distributed across the entire genome (bac, alp, pil1 locus, pepS8, fBsB, capsule locus, hylB, abiG-I/-II plus Q8DZ34, pil2 locus, nss plus srr plus rogB2 and rgfC/A/D/B). Ubiquitous, conserved genes (e.g. cylD/cylE) were included to be used as species markers and controls. Furthermore, probes for highly mobile plasmid-borne resistance genes such as erm(A), erm(B), erm(C), tet(M), emrB/qacA were also included (Aracil, et al., 2002; Betriu, et al., 2003; Uh, et al., 2001). These genes are not specific to GBS, but they are found in some isolates. They can be used to distinguish individual, related isolates, rather than for a definition of distinct strains. A total of 448 isolates of GBS was selected and examined for the present work. Among them were isolates from severe diseases, such as sepsis and meningitis, isolates from local infections as well as isolates from asymptomatic/healthy carriers. For comparison, isolates from veterinary medicine (from cases of bovine mastitis) and human isolates from a geographically distant region (Trinidad and Tobago) were genotyped. For 36 selected isolates with representative hybridization patterns, parallel typing was performed using a second method, multilocus sequence typing (MLST) (Jones, et al., 2003). Hybridization patterns on the Microarray could thus be linked to this already established typing system. With the array based GBS typing isolates could be divided into 76 different strains or 'hybridization profiles', HP. The classification with both methods is based on the absence or presence of individual genes or gene clusters or their allelic variants. Similar isolates were lumped together. The eBURST algorithm was used to group strain-specific patterns into groups of related strains illustrating the population structure and describing the genetic diversity. Groups of similar hybridization patterns largely correlate with the clonal complexes (CC) defined by MLST. While isolates with identical or similar hybridization profiles belong to the same CC, isolates with a similar MLST profile can show different hybridization patterns. It has also often been observed that otherwise similar strains differ from each other in individual traits, e.g. capsule genes, alp or pili genes, and that these genes vary independently of one another. In addition, some ubiquitous genes/gene clusters, which are always localized at the same position in the published genomic sequences, show numerous different alleles and related strains (that belong to one clonal complex) might differ in the presence of one allele. Alleles are thus not linked to clonal complexes, but rather randomly distributed. An explanation of this phenomenon could be a frequent occurrence of recombination events or horizontal gene transfers. A convergent evolution could also be discussed as an alternative explanation. A similarity of hybridization profiles does not necessarily mean a direct phylogenetic relationship between the isolates in the sense of being derived from a direct common ancestor. Typing both the DNA microarray and the MLST cannot reflect the 'true' phylogenesis, evolutionary history and origin. Assuming frequent recombination i.e., random events, the MLST profiles as well as the hybridization patterns can be used as genetic fingerprints, allowing a comparison of isolates, but no conclusions about their phylogeny and origin. The investigated GBS isolates were classified into five clonal complexes (CC19, CC23, CC26, CC103, CC130) with very different relative abundances indicating differences in the population structure with regard to geographic origin and host organisms. German medical isolates were mainly assigned CC19. Caribbean medical isolates mostly were assigned to CC19 and CC23. Bovine isolates usually belonged to CC19 and CC103. Among human isolates, CC103 was rare. In the present work, CC19 was the most abundant and the genetically most inhomogeneous CC. Several different clusters that previously been regarded as CCs (CC1, CC17, CC19 and CC22) have recently been merged to CC19 by the eBURST algorithm since MLST profiles of missing links between the CCs have been identified. Unfortunately, eBURST cannot distinguish whether two MLST types are linked by true common ancestors or by hybrid or chimera strains originating from horizontal gene transfers or hybridization events. Since these complexes within CC19 have clearly distinguishable hybridization patterns, they have been referred to herein as CC19/01, CC19/17, CC19/19 and CC19/22, and we assume that they are linked by hybridizations or gene transfers rather than by shared ancestry. Few differences were found between isolates from different origins. The virulence factor scpB was found in 412 of 418 human isolates (98.6%), but only in 10 of 21 bovine isolates (48%). Furthermore, it was observed that invasive isolates are less likely to carry abiGI-/II and Q8DZ34, but are more likely to have pil1 locus, fbsB (515) and capsule type III as well as pil2b, nss/srr and rgf (COH1 like). Some of these markers appear together in CC19/17 strains, which are often observed in invasive disease. speM (exotoxin M) was also investigated. It was detected only in seven isolates. Contrarily, the cyl (beta-hemolysin) operon was found in all isolates. Thus, it detection is not helpful for a prediction of the virulence of a given isolate. No single factor could be identified that allowed a definitive distinction between invasive isolates and carrier isolates. Probably, the virulence of an isolate does not depend on the presence or absence of one particular gene. In this context, it would be important to investigate the expression in vivo of the various putative virulence factors as well as the allelic variants of the factors and of their associated regulatory genes. Macrolide-/clindamycin resistance genes erm(A), erm(B) and erm(C) can also be detected by the microarray. None of these genes was identified in any of the Caribbean isolates. Within the German GBS population, erm(B) was most frequently observed. The genes erm(A) and erm(C) were rare in human isolates, and they were not found in bovine isolates. The tetracycline resistance gene tet(M) was observed frequently in human isolates but only very rarely in veterinary isolates. With the genotyping method that was developed during the present work, further studies can be carried out to study regional and host-specific differences in the GBS population. For future investigations, the described typing method could further be refined. For example, the 11 selected typing markers on the microarray can be combined with those from MLST to one comprehensive marker system. In addition, it is also possible to use genes on mobile genetic elements such as resistance genes (erm, cad, mer or tet) to prove or to rule out the identity of related isolates in vitro or in silico. In our study, it was shown that the DNA microarray provides stable and reproducible results that are comparable to those of other typing methods and genome sequencing. However, since the DNA microarray offers a much more uncomplicated and faster procedure, which also results in lower costs, it is more suitable to a routine setting.:1. Einleitung 5 1.1 Gegenstand der Untersuchung 5 1.2 Entdeckungsgeschichte 6 1.3 Klinische Bedeutung 7 1.4 Veterinärmedizinische Bedeutung 9 1.5 Stand der Forschung mit Hinblick auf Typisierung von GBS 9 1.6 Ziele der vorliegenden Untersuchung 10 1.7 Vorgehensweise 11 1.7.1 Untersuchungsmaterial 11 1.7.2 Untersuchungsmethode 11 1.8 Zur Typisierung ausgewählte Marker 12 2. DNA Microarray-Based Typing of Streptococcus agalactiae Isolates 15 2.1 Abstract 16 2.2 Introduction 17 2.3 Materials and methods 18 2.3.1 Bacterial isolates 18 2.3.2 Ethics statement 18 2.3.3 Preparation of genomic DNA 19 2.3.4 MLST 19 2.3.5 Microarray design and protocol optimization 19 2.3.6 Microarray procedures 20 2.3.7 eBURST 21 2.4 Results 22 2.4.1 Typing GBS by MLST 22 2.4.2 Genotyping GBS by microarray hybridization 22 2.4.3 Population structure 25 2.4.4 Detection of antibiotic resistance markers 25 2.4.5 Detection of heavy metal resistance markers 26 2.5 Discussion 27 2.6 Acknowledgments 30 2.7 References 31 2.8 Tables and figures 33 3. Zusammenfassung 45 3.1 Zusammenfassung 45 3.2 Summary 49 4. Korrespondenz mit dem Editor 53 4.1 Hinweise des Editors und der Gutachter 53 4.2 Antworten an den Editor 56 4.3 Endgültige Annahme 58 Anhang 61

Gene expression can serve as a powerful predictor for disease progression and other phenotypes. Consequently, microarrays, which capture gene expression genome-wide, have been used widely over the past two decades to derive biomarker signatures for tasks such as cancer grading, prognosticating the formation of metastases, survival, and others. Each of these signatures was selected and optimized for a very specific phenotype, tissue type, and experimental set-up. While all of these differences may naturally contribute to very heterogeneous and different biomarker signatures, all cancers share characteristics regardless of particular cell types or tissue as summarized in the hallmarks of cancer. These commonalities could give rise to biomarker signatures, which perform well across different phenotypes, cell and tissue types. Here, we explore this possibility by employing a network-based approach for pan-cancer biomarker discovery. We implement a random surfer model, which integrates interaction, expression, and phenotypic information to rank genes by their suitability for outcome prediction. To evaluate our approach, we assembled 105 high-quality microarray datasets sampled from around 13,000 patients and covering 13 cancer types. We applied our approach (NetRank) to each dataset and aggregated individual signatures into one compact signature of 50 genes. This signature stands out for two reasons. First, in contrast to other signatures of the 105 datasets, it is performant across nearly all cancer types and phenotypes. Second, It is interpretable, as the majority of genes are linked to the hallmarks of cancer in general and proliferation specifically. Many of the identified genes are cancer drivers with a known mutation burden linked to cancer. Overall, our work demonstrates the power of network-based approaches to compose robust, compact, and universal biomarker signatures for cancer outcome prediction.

Poverty and long distances are two reasons why some people in the third world countries hasdifficulties seeking medical help. A solution to the long distances could be if the medical carewas more mobile and diagnostically tests could be performed on site in villages. A new pointof-care test based on a small blood shows promising results both in run time and mobility.However, the method still needs more advanced equipment for analysis of the resultingmicroarray. This study has investigated the potential to perform the analysis within asmartphone application, performing all steps from image capturing to a diagnostic result. Theproject was approach in two steps, starting with implementation and selection of imageanalysis methods and finishing with implementing those results into an Android application.A final application was not developed, but the results gained from this project indicates that asmartphone processing power is enough to perform heavy image analysis within a sufficientamount of time. It also imply that the resolution in the evaluated images taken with a Nexus 6together with an external macro lens most likely is enough for the whole analysis, but furtherwork must be done to ensure it.

Bladder cancer is a potentially curable malignancy; however in regards to the state of current therapy regimens, a plateau has been reached in both the non-muscle and muscle invasive types. To obtain effective treatment, and consequently a decreased mortality, it has become imperative to test and understand aspects affecting therapy response. The aim of this thesis is to illustrate a better understanding of clinical factors affecting therapy response using new drug combinations and new tumor markers alongside established risk criteria. In Paper I we reported the 5 year follow up from a multicenter, prospectively randomized study and we evaluated the 5-year outcomes of BCG alone compared to a combination of epirubicin and interferon-a2b in the treatment of patients with T1 bladder cancer. Treatment, tumor size and tumor status at second resection were independent variables associated with recurrence. Concomitant Cis was not predictive of failure of BCG therapy. Independent factor for treatment failure was remaining T1 stage at second resection. In Paper II &III we investigated the validity of emmprin, survivin and CCTα proteins as biomarkers for response and survival before neoadjuvant cisplatin chemotherapy. Bladder tumor specimens were obtained before therapy from a total of 250 patients with T1-T4 bladder cancer enrolled in 2 randomized trials comparing neoadjuvant chemotherapy before cystectomy with a surgery only arm. Protein expression was determined by immunohistochemistry (IHC). Patients in the chemotherapy cohort with negative emmprin and CCTα expression had significantly better overall survival (OS) than those with positive expression. In Paper IV primary end point was examining STMN1 as prognostic factor in bladder cancer.  Analysis was performed on three bladder cancer patient cohorts using IHC, western blot and a bladder cancer cell line. High levels of STMN1, expression correlated to shorter disease-specific survival and the growth and migration of the cells were significantly reduced when transfecting the cells with STMN1 siRNA. Conclusion Risk assessment and predictors of outcomes could help in individualized treatment and follow up.  Biomarkers will become more important for treatment choices in bladder cancer management.