Urinary tract infection (UTI) is one of the most serious health problems worldwide. It accounts for a million hospital visits annually in the United States. Among the many uropathogenic bacteria, uropathogenic Escherichia coli (UPEC) is the most common causative agent of UTI. However, not all E. coli that inhabit the urinary tract can cause UTI. Some of them thrive for long periods of time in the urinary bladder without causing overt symptoms of infection. This carrier state is called asymptomatic bacteriuria (ABU). E. coli ABU isolates can live in the host without inducing host response due to deletions, insertions and point mutations in the genome leading to the attenuation of virulence genes. They therefore behave in the same way as commensals. Since bacteria that inhabit the urinary tract are said to originate from the lower intestinal tract and ABU behave in a similar way as commensals, this study compared various phenotypic and genotypic characteristics of ABU and commensal E. coli fecal isolates. The two groups did not show a strict clustering with regards to phylogenetic lineage since there appears to be overlaps in their distribution in some clonal complexes. In addition, it was observed that the UPEC virulence genes were more frequently inactivated in ABU than in fecal isolates. Hence, ABU tend to have less functional virulence traits compared to the fecal isolates. The ABU model organism E. coli 83972 which is known not only for its commensal behavior in the urinary bladder but its ability to outcompete other bacteria in the urinary tract is currently being used as prophylactic treatment in patients who have recurrent episodes of UTI at the University Hospital in Lund, Sweden. The pilot studies showed that upon deliberate long-term colonization of the patients with E. coli 83972, they become protected from symptomatic UTI. In this study, the phenotypic and genotypic characteristics of eight re-isolates taken from initially asymptomatically colonized patients enrolled in the deliberate colonization study who reported an episode of symptoms during the colonization period were investigated. Two out of the eight re-isolates were proven to be a result of super infection by another uropathogen. Six re-isolates, on the other hand, were E. coli 83972. The urine re-isolates confirmed to be E. coli 83972 were phenotypically heterogeneous in that they varied in colony size as well as in swarming motility. Four of these re-isolates were morphologically homogenous and similar to the parent isolate E. coli 83972 whereas one of them appeared phenotypically heterogenous as a mixture of smaller and normal-sized colonies. Still another re-isolate phenotypically resembled small colony variants. Meanwhile, three of the six re-isolates did not differ from the parent isolate with regards to motility. On the other hand, three exhibited a markedly increased motility compared to the parent isolate. Transcriptome analysis demonstrated the upregulation of a cascade of genes involved in flagellar expression and biosynthesis in one of the three motile re-isolates. However, upon further investigation, it was found out that the expression of flagella had no effect on bacterial adhesion to host cells in vitro as well as to the induction of host inflammatory markers. Thus, this implies that the increased motility in the re-isolates is used by the bacteria as a fitness factor for its benefit and not as a virulence factor. In addition, among the various deregulated genes, it was observed that gene regulation tends to be host-specific in that there is no common pattern as to which genes are deregulated in the re-isolates. Taken together, results of this study therefore suggest that the use of E. coli 83972 for prophylactic treatment of symptomatic UTI remains to be very promising. / Harnwegsinfektionen (HWI) sind weltweit ein ernstes Gesundheitsproblem, auf welches allein in den USA jährlich ca. eine Million Krankenhausbesuche entfallen. Innerhalb der Gruppe uropathogener Bakterien stellen die uropathogenen Escherichia coli (UPEC) die wichtigsten Verursacher akuter Harnwegserkrankungen dar. Interessanterweise führen nicht alle E. coli Varianten, die den Harnweg besiedeln, zwangsläufig zu HWI. Einige von ihnen sind in der Lage, die Harnblase über einen langen Zeitraum zu kolonisieren ohne Symptome einer HWI auszulösen. Dieses Phänomen wird als asymptomatische Bakteriurie (ABU) bezeichnet. Die Eigenschaft von E. coli ABU-Isolaten innerhalb des Wirtsorganismus leben zu können, ohne eine deutliche Wirtsabwehrreaktion hervorzurufen, ist unter anderem bedingt durch Deletionen, Insertionen und Punktmutationen im bakteriellen Genom und der daraus resultierenden Inaktivierung einiger Virulenzgene. Ihre Lebensweise ist daher mit der kommensaler Organismen vergleichbar. Da die den Harnweg besiedelnden Bakterien mit hoher Wahrscheinlichkeit ihren Ursprung im unteren Darmtrakt haben und sich die ABU-Isolate ähnlich der Kommensalen verhalten, wurden in dieser Arbeit zahlreiche phäno- und genotypische Charakteristika von ABU-Isolaten mit denen kommensaler E. coli-Fäkalisolate verglichen. Für diese beiden Gruppen konnte hinsichtlich ihrer phylogenetischen Abstammung keine strikte Clusterbildung festgestellt werden, da ihre Verteilung in einigen klonalen Komplexen Überlappungen aufwies. Es zeigte sich jedoch, dass die UPEC-Virulenzgene in den ABU-Isolaten häufiger inaktiviert vorlagen als in den Fäkalisolaten. Demzufolge scheinen die ABU-Isolate weniger funktionale Virulenzeigenschaften zu besitzen als die Fäkalisolate. Der Modell-ABU Stamm E. coli 83972 ist sowohl für sein kommensales Verhalten in der menschlichen Blase bekannt, als auch für seine Fähigkeit, andere Bakterien aus dem Harntrakt zu verdrängen. Er wird gegenwärtig am Universitätsklinikum in Lund (Schweden) als prophylaktisches Therapeutikum bei der Behandlung von Patienten mit rezidivierenden HWI eingesetzt. In Pilotstudien konnte gezeigt werden, dass eine vorsätzliche Langzeit-Kolonisierung von Patienten mit E. coli 83972 zum Schutz vor symptomatischen HWI führt. In der vorliegenden Arbeit wurden die phäno- und genotypischen Charakteristika von acht Patienten-Reisolaten dieser Langzeitstudie untersucht. Die Reisolate wurden aus zunächst asymptomatisch kolonisierten Patienten isoliert, die allerdings im Verlauf der Langzeit-Kolonisierung über eine Reihe von Symptomen klagten. Zwei dieser acht Reisolate waren nachweislich das Resultat einer Superinfektion mit einem anderen uropathogenen Bakterium. Die restlichen sechs Reisolate konnten jedoch als E. coli 83972 identifiziert werden. Für diese sechs Reisolate war eine phänotypische Heterogenität zu beobachten, die sich zum einen in variierender Koloniegröße, zum anderen in unterschiedlichem Schwärmverhalten zeigte: Vier der Reisolate entsprachen morphologisch dem Ausgangsstamm E. coli 83972, wohingegen ein Reisolat phänotypisch abweichend als Mischung von kleineren und normal-großen Kolonien in Erscheinung trat. Ein weiteres Reisolat ähnelte phänotypisch sogenannten „Small Colony Variants“. Das Schwärmverhalten betreffend unterschieden sich indessen drei der sechs Reisolate vom Ausgangsstamm. Sie zeigten im Vergleich eine erhöhte Motilität. In einem dieser drei motilen Reisolate konnte mittels Transkriptomanalyse die hochregulierte Expression einer Reihe von Genen, welche für die Flagellenexpression und -biosynthese verantwortlich sind, aufgezeigt werden. Weiterführende in vitro-Untersuchungen ergaben jedoch, dass diese erhöhte Flagellenexpression weder einen verstärkenden Effekt auf die bakterielle Adhäsion an die Wirtszellen hat, noch in der Wirtszelle die Bildung von Entzündungsmarkern induziert. Dieses Ergebnis impliziert, dass die erhöhte Motilität der Reisolate als Fitnessfaktor und nicht als Virulenzfaktor zu betrachten ist. Ferner führte die genauere Analyse der deregulierten Gene zu der Annahme, dass die Genregulation in den Reisolaten wirtsspezifisch ist, da sich kein übereinstimmendes Muster bezüglich der Deregulation abzeichnete. Unter Berücksichtigung aller Ergebnisse dieser Studie lässt sich abschließend sagen, dass die Verwendung des E. coli Stammes 83972 als prophylaktisches Therapeutikum bei der Behandlung von symptomatischen HWI weiterhin als sehr vielversprechend angesehen werden kann.

Chromosomally determined arsenate resistance in Escherichia coli is well characterized. Little, however, is known about chromosomally determined arsenite resistance in E. coli. Accordingly, spontaneous arsenite resistant mutants were selected in a plasmid free strain of E. coli. One mutant strain was analyzed in detail, genetically and physiologically. The mutation confering arsenite resistance was shown to be a single gene mutation. Genetic mapping studies using conjugation and transduction showed that the mutation was closely linked to the ilv region of the E. coli map.

DNA interstrand cross-links are a form of genomic damage that cause a block to replication and transcription of DNA in cells and cause lethality if unrepaired. Chemical agents that induce cross-links are particularly effective at inactivating rapidly dividing cells and, because of this, have been used to treat hyperproliferative skin disorders such as psoriasis as well as a variety of cancers. However, evidence for the removal of cross-links from DNA as well as resistance to cross-link-based chemotherapy suggests the existence of a cellular repair mechanism. Characterizing the pathways involved in DNA interstrand cross-link repair has been challenging due to the inherent structure of the damage as it precludes the use of an undamaged, complementary strand of DNA as a template for repair. A number of models of cross-link repair have been proposed based on the identification of hypersensitive repair mutants as well as biochemical evidence that specific repair enzymes are capable of incising cross-linked structures from DNA. Together, these models have suggested the involvement of multiple repair pathways--such as nucleotide exicision repair, translesion synthesis, recombination of double-strand breaks, and base excision repair--operating in sequential steps to correct the damage. Most of the studies from which these models arose are complicated by the fact that cross-linking agents induce multiple forms of damage or they lack in vivo confirmation of how the repair phenomenon occurs in organisms. In this study, I use Escherichia coli as a model organism to examine the involvement of the aforementioned pathways in DNA interstrand cross-link repair in vivo. This organism was useful in early cross-link studies and, with its highly conserved repair processes, maintains the potential for delineating how cross-links are removed in higher organisms. In Chapter I, I introduce background information on different cross-linking agents, the complications of studying cross-link repair, and the candidate repair pathways that have been implicated to date. In Chapter II I demonstrate that there is a limited involvement of the nucleotide excision repair helicase, translesion polymerases, and double-strand break repair enzymes through survival analysis of cells defective in these proteins. For this analysis, I use 8-methoxypsoralen plus UVA as a cross-linking agent and angelicin plus UVA as a monofunctional comparator. The observation that uvrD mutants-- defective in helicase II of nucleotide excision repair--were nearly as resistant to 8-methoxypsoralen-induced damage as wild type cells led me to examine the incision rate of cross-links from endogenous plasmid DNA. Surprisingly, cross-links were not efficiently removed from DNA in uvrD mutants relative to wild type cells. These seemingly contradictory results were rectified when I quantified cross-link formation in cell cultures and revealed that as few as one cross-link per chromosome can inactive wild type cells, a lethal quantity that is lower than what has been previously reported. Taken together, these observations suggest that although cross-links are incised in wild type cells, repair is still not a highly productive event in E. coli. In Chapter III I examine the involvement of the base excision repair pathway in cross-link repair and demonstrate that Nth and Fpg Glycosylases, Xth and Nfo AP-Endonucleases sensitize Escherichia coli to psoralen-induced DNA damage. This is shown by comparative survival analysis in angelicin plus UVA and 8-methoxypsoralen plus UVA treatment whereby nth-, fpg-, and xth-mutants are each more resistant than wild type cells to either treatment. This suggests that when these gene products are present they impact the production or removal of monoadducts. nfo-mutants were different in that the cells were only hyperresistant to 8-methoxypsoralen monoadducts and cross-links, either implying that the Nfo enzyme interacts specifically with psoralen monoadducts rather than angelicin monoadducts or that the enzyme impedes cross-link removal. Finally, in Chapter IV a summary of the results is provided as well as future directions that may be explored following this study.

Die Forschungsergebnisse der letzten Jahre liefern immer mehr Hinweise darauf, dass eine klare Unterscheidung von Fitness- und Virulenzfaktoren in vielen Fällen, insbesondere bei extraintestinal pathogenen Escherichia coli, nicht möglich ist. So lässt sich auch bei Harnwegsinfektionen verursachenden E. coli den bakteriellen und teils stammspezifischen Faktoren oftmals nicht eindeutig eine typische Virulenz- oder Fitness-assoziierte Funktion zuordnen. Zudem werden in neueren Studien immer häufiger atypische uropathogene Isolate von E. coli beschrieben, die in ihrem „Virulenzrepertoire“ deutlich von typischen uropathogenen E. coli (UPEC) abweichen, da sie keine klassischen UPEC-Virulenzfaktoren aufweisen. In dieser Arbeit wurden daher Virulenzeigenschaften typischer als auch atypischer UPEC untersucht. Der Effekt eines bestimmten bakteriellen Faktors auf den Wirtsorganismus wird teilweise indirekt durch sekundäre Modifikation bedingt. Dies offenbart sich beispielsweise am Autotransporterprotein AIDA-I, dessen Konformation durch posttranslationale Glykosylierung stabilisiert wird, wodurch es seine Funktionalität als Adhäsin erhält. Da bisherige Studien zum AIDA-I homologen Autotransporterprotein Antigen 43 (Ag43) auf der Analyse von künstlich glykosyliertem Protein basieren, lag ein Schwerpunkt dieser Arbeit auf der Untersuchung der natürlichen Glykosylierung von Ag43 in UPEC Stamm 536. Es zeigte sich, dass beide Ag43-Varianten von E. coli Stamm 536 natürlicherweise glykosyliert vorliegen, der Grad der Glykosylierung jedoch wesentlich geringer ausfällt als bei natürlich glykosyliertem AIDA-I. Inwieweit die natürliche Glykosylierung von Ag43 zu dessen Funktionalität beiträgt, kann erst durch die Identifizierung der für die Ag43-Glykosylierung verantwortlichen Glykosyltransferase geklärt werden. Die in silico-Analyse des Genoms von UPEC Stamm 536 für potentielle Glykosyltransferasen von Ag43 lieferte neun Kandidatengene. Die Gene wurde teils im Wildtyp-Hintergrund, teils im rfaH-negativen Hintergrund von E. coli Stamm 536 deletiert und die Mutanten im Anschluss phänotypisch charakterisiert. Die Deletion der Kandidatengene waaF, waaG und waaQ, die für Glykosyltransferasen des LPS-Biosynthesesystems kodieren, führte zu den deutlichsten Unterschieden in Bezug auf Motilität, Curli/Zellulose-Produktion, Hämolyseaktivität und Expression von Typ 1 Fimbrien. Der Einfluss des „knock-out“ der Kandidatengene auf die Glykosylierung von Ag43 muss in weiterführenden Studien untersucht werden. Zur Charakterisierung des uropathogenen Virulenzpotentials verschiedener E. coli Stämme in vivo hat sich in den letzten Jahren das murine Modell der aufsteigenden Harnwegsinfektion etabliert. Mit Hilfe dieses Modells wurden in der vorliegenden Arbeit sowohl spezifische Deletionsmutanten prototypischer UPEC als auch atypische E. coli Harnwegsisolate bezüglich ihrer Urovirulenz getestet und verglichen. Bei der Untersuchung der klassischen UPEC lag der Fokus auf der möglichen Urovirulenzmodulation durch die folgenden spezifischen Faktoren: dem Autotransporterprotein Ag43, dem „Response regulator“ UvrY, dem Polyketid Colibactin sowie dem Exopolysaccharid poly-β-1,6-N-Acetylglucosamin (PGA). Für Ag43 war bei der Etablierung einer Harnwegsinfektion keine eindeutige Funktion feststellbar. Es ist jedoch denkbar, dass Ag43 zur Langzeitpersistenz im Harnwegstrakt beitragen kann, was in weiteren Studien belegt werden sollte. Die Expression von UvrY in der natürlichen uvrY-Deletionsmutante UPEC Stamm 536 ließ keine Erhöhung des Urovirulenzpotentials im Mausmodell erkennen. In diesem Zusammenhang konnte allerdings gezeigt werden, dass die Expression des Genotoxins Colibactin in UPEC Stamm 536 dessen Virulenz signifikant herabsetzte. Die Untersuchungen zur Relevanz des Exopolysaccharids PGA belegen deutlich, dass PGA für die Langzeitpersistenz von E. coli im murinen Harnwegstrakt benötigt wird. Für die initiale Kolonisierung scheint PGA hingegen keine Bedeutung zu haben. Für atypische UPEC Isolate, die Charakteristika von STEC und EAEC zeigen und sich in ihrem Virulenzmuster deutlich von prototypischen UPEC unterscheiden, ließ sich im murinen Modell der aufsteigenden Harnwegsinfektion, verglichen mit dem UPEC Modellorganismus 536, ein ähnliches, teils sogar erhöhtes uropathogenes Virulenzpotential nachweisen. Die Ergebnisse der Arbeit untermauern somit die heutige Vorstellung bezüglich der Entwicklung und Etablierung einer Harnwegsinfektion, dass verschiedene E. coli Stämme unterschiedliche (Kontroll-) Mechanismen entwickelt haben, um erfolgreich den Harnwegstrakt kolonisieren und eine Infektion auslösen zu können. Zudem weisen sie darauf hin, dass diese Fähigkeit nicht auf Isolate typischer phylogenetischer UPEC Entwicklungslinien beschränkt und auf das Vorhandensein charakteristischer UPEC Virulenzfaktoren angewiesen ist. / Research findings over the last years indicate that in many cases, including extraintestinal pathogenic Escherichia coli, a clear distinction between fitness and virulence factors is not possible. Accordingly, the classical distinction of often strain-specific virulence- and fitness-related traits of uropathogenic E. coli (UPEC) can often not be made. Furthermore, recent studies describe atypical UPEC isolates. These isolates remarkably differ in their “virulence repertoire” compared to typical UPEC, because they lack classical UPEC-related virulence factors. Therefore, the aim of the present study was the investigation of virulence properties of typical as well as atypical UPEC strains. The effect of a certain bacterial factor upon the host organism is in part indirectly influenced by secondary modifications. For instance, the conformation of the autotransporter protein AIDA-I is stabilized by posttranslational glycosylation which in turn confers its functionality as an adhesin. Prior studies on the AIDA-I homologous autotransporter protein antigen 43 (Ag43) are based on the analysis of the artificially glycosylated protein. Thus, a key aspect of the current work was to elucidate the naturally occurring glycosylation of Ag43 in UPEC strain 536. For both Ag43 variants of E. coli 536 natural glycosylation was detected. However, Ag43 was less glycosylated than naturally glycosylated AIDA-I. The future identification of the glycosyltransferase responsible for natural glycosylation of Ag43 will help to determine the impact of this posttranslational modification on the functionality of Ag43. In silico analysis of the UPEC strain 536 genome regarding potential glycosyltransferases of Ag43 revealed nine candidate genes. Corresponding deletion mutants of the identified genes were constructed in part in the wild type strain background and in part in the rfaH-negative background of UPEC 536. The most prominent differences concerning motility, curli/cellulose production, hemolytic activity and expression of type 1 fimbriae were observed upon deletion of the genes waaF, waaG or waaQ coding for glycosyltransferases of the LPS biosynthesis pathway. The impact of the deleted candidate genes on the glycosylation of Ag43 has to be further investigated. In recent years the murine model of ascending urinary tract infection was established to characterize the uropathogenic potential of E. coli strains in vivo. By means of this model the uropathogenic potential of different specific “knock-out” mutants of prototypic UPEC strains as well as of atypical E. coli urinary tract isolates was tested and compared. The analysis of the impact of specific factors on the uropathogenic potential of classical UPEC strains focused on the autotransporter protein Ag43, the response regulator UvrY, the genotoxin colibactin, and the exopolysaccharide poly-β-1,6-N-acetylglucosamine (PGA). Ag43 did not exhibit a distinct function during the establishment of urinary tract infection in mice. However, it is conceivable that Ag43 can contribute to long-term persistence in the urinary tract, which should be covered in further studies. Expression of UvrY in the natural uvrY-negative UPEC strain 536 did not increase the uropathogenic potential. However, expression of the genotoxin colibactin significantly reduced the urovirulence of UPEC strain 536. The exopolysaccharide PGA was shown to contribute to long-term persistence of UPEC in the murine model of urinary tract infection. For the initial colonization of the urinary tract, PGA seems to be dispensable. The atypical UPEC isolates investigated in this study display typical characteristics of STEC and EAEC and differ significantly in their virulence gene content compared to prototypic UPEC strains. Nevertheless, in the murine model of ascending UTI many atypical UPEC isolates exhibited a comparable and sometimes even increased uropathogenic potential relative to UPEC model strain 536. The results of this work support the current idea regarding the development and establishment of a urinary tract infection that different E. coli strains have evolved diverse (control-) mechanisms to successfully colonize the urinary tract and provoke an infection. In addition, the findings point out that the ability to cause a urinary tract infection is not limited to phylogenetic lineages of classical UPEC isolates and the presence of characteristic UPEC virulence traits.

E. coli Nissle 1917 (EcN) zählt durch seine fast hundertjährige Nutzung als Arzneimittel und aufgrund der weitreichenden Forschung während der letzten Jahrzehnte mittlerweile zu einem der am besten untersuchten Probiotika. EcN wird als Medikament zur Remissionserhaltung von Patienten mit Kolitis, bei chronischer Verstopfung und bei Durchfall von Kleinkindern eingesetzt. Der enteroaggregative – hämorrhagische - E. coli (EAHEC) mit dem Serotyp O104:H4 war 2011 in Deutschland für den bisher größten EHEC-Ausbruch seit Beginn der Aufzeichnungen verantwortlich. Es fehlt bis zum heutigen Tage immer noch an effektiven Möglichkeiten einer Infektionsprophylaxe oder einer Behandlung der Erkrankung. Ein alternatives Therapeutikum wird daher dringend benötigt. In dieser Arbeit wurden die antagonistischen Effekte von EcN auf pathogene E. coli Stämme wie dem EHEC Stamm EDL933 oder klinischen EAHEC O104:H4 Isolaten untersucht. Es wurden die Auswirkungen von EcN auf die Adhäsion an humane Epithelzellen, das Wachstum und die Shiga Toxin Produktion der pathogenen Stämme untersucht. Zusätzlich wurde die Resistenz von EcN gegenüber Shiga Toxin Phagen nachgewiesen. Zunächst wurde die Adhäsionseffizienz der verschiedenen E. coli Stämme bestimmt. Der am schlechtesten an die humanen Epithelzelllinien Caco-2 und LS-174T adhärierende Stamm war EcN. Dies ist insofern überraschend, da von Probiotika erwartet wird, besser als Pathogene an Epithelzellen zu adhärieren. Dem ungeachtet konnte jedoch gezeigt werden, dass EcN die Adhäsion von zwei EAHEC O104:H4 Isolaten, des nahe verwandten enteroaggregativen E. coli (EAEC) Stammes 55989 und des enterohämorrhagischen (EHEC) E. coli Stammes O157:H7 EDL933 an beide Zelllinen hemmt. Die von EcN produzierten Mikrozine M und H47 konnten hier für einen Teil des beobachteten anti-adhäsiven Effektes von EcN auf die pathogenen E. coli Stämme verantwortlich gemacht werden. Die Mikrozine wurden hier als einzige Substanz, die das Wachstum der pathogenen E. coli Stämme beeinflusst, identifiziert. Einer der wichtigsten Virulenzfaktoren von EAHEC und EHEC Stämmen ist das Shiga Toxin. In dieser Arbeit konnte gezeigt werden, dass EcN die Shiga Toxin Produktion der am häufigsten auftretenden EHEC Stämme (´Big Five´: O157:H7, O26:H11, O103:H2, O111:H-, O145:H25) und der klinischen Isolate von EAHEC O104:H4 im Zellkulturmedium DMEM hemmt. Auffällig war, dass die Stx1 Produktion von EHEC O103:H2 und O111:H- nicht nur von EcN, sondern auch von E. coli K-12 Stamm MG1655, gehemmt wurde, im Gegensatz zur EcN-spezifischen Blockierung der Stx2-Produktion in den Serotypen O104:H4, O26:H11, O145:H25. Die Reduktion der Stx-Produktion in EAHEC O104:H4 TY3730 und TY3456, sowie EHEC O26:H11 war zum Teil von der Mikrozinproduktion abhängig. Diese hatte jedoch keinen Einfluss auf die Stx-Produktion in EHEC O157:H7 EDL933 und EHEC O145:H25. Bei Verwendung von LB-Medium zeigte sich im Gegensatz zum DMEM-Medium keine Mikrozin-Abhängigkeit der Toxinproduktion bei den EAHEC Isolaten TY3730 und TY3456. Die Toxinproduktion von EHEC EDL933 wurde ebenfalls nicht durch die Deletion der Mikrozin-Gene in EcN beeinflusst. Studien der Toxinproduktion in SCEM-Medium zeigten ebenfalls eine EcN-Dosisabhängige Reduktion der Stx-Produktion in Co-Kultur. Um den Mechanismus der Hemmung der Stx-Produktion zu untersuchen, wurden Versuche mit der EcN-Mutante EcN::luxS durchgeführt. Diese Deletion des AI-2 ´Quorum sensing´ Moleküls in EcN hatte allerdings keinen Einfluss auf die Hemmung der Stx-Produktion. Der Einsatz von Acetat führte, im Gegensatz zu publizierten Ergebnissen, nicht zu einer Reduktion der Stx-Produktion. Auch eine Beeinflussung der Lyse der EHEC-Bakterien, oder der Verminderung der Sekretion von Shiga Toxin durch EcN, konnte widerlegt werden. Zur Untersuchung der Stx-Expression wurde ein Assay mit einem biolumineszenten C-P (Chromosom-Plasmid) Reporter System etabliert. Damit konnte die Shiga Toxin Expression im Stammhintergrund EHEC EDL933 in Echtzeit untersucht werden. Hier wurde wiederum eine Reduktion der Shiga Toxin Expression in Co-Kultur mit EcN erfolgreich nachgewiesen. In weiteren Versuchen konnte gezeigt werden, dass EcN nicht nur die Shiga Toxin Produktion von nicht-induzierten EAHEC Bakterien, sondern auch in mit Mitomycin C induzierten Bakterien hemmt. Als wichtiger Sicherheitsaspekt einer Behandlung mit EcN wurde die Resistenz von EcN gegenüber Shiga Toxin Phagen untersucht. Die Infektion der Bakterien wurde hierbei mit stx-spezifischer PCR, Phagen-Plaque-Assay, Stx-ELISA und K+-Efflux Assay untersucht. Es konnte durch diese verschiedenen Methoden erfolgreich gezeigt werden, dass EcN nicht durch Shiga Toxin Phagen infiziert wird. Als möglicher Resistenzmechanismus kommt hier eine Mutation vom Phagenrezeptor LamB in Frage, was jedoch noch bestätigt werden muss. Zusammenfassend wurden in dieser Arbeit wichtige antagonistische Effekte von EcN auf pathogene E. coli Stämme untersucht, die als Grundlage von neuen und dringend benötigten Behandlungen von EHEC-Infektionen dienen können. / Due to extensive studies in the last decades and its centennial application as a pharmaceutical, E. coli Nissle 1917 (EcN) is among the best characterized probiotics. EcN is used as remedy for remission maintenance of ulcerative colitis, chronic obstipation and diarrhea in children. The enteroaggregative – haemorrhagic - E. coli (EAHEC) strain O104:H4 was responisible for one of the biggest outbreaks of EHEC recorded so far, that took place in Germany in 2011. Currently, there is no effective prophylaxis or treatment available for EHEC infections in humans. Therefore, alternative therapeutics are desperately needed. The antagonistic effects of EcN on pathogenic E. coli strains like the EHEC O157:H7 strain EDL933 or clinical isolates of EAHEC O104:H4 were investigated in this study. The influence of EcN on adhesion to human epithelial cell lines, the growth and the Shiga toxin production of pathogenic strains were analysed. Furthermore, the resistence of EcN against Shiga toxin phages was proven. Initially, the adhesion efficiency of EcN and pathogenic E. coli strains were determined in monocultures. EcN showed the lowest number of adhering bacteria to Caco-2 and LS-174T cells. This was insofar surprising, since probiotics are expected to adhere more efficiently to epithelial cells than pathogens. Regardless of this fact, it could be shown that EcN is inhibiting the adhesion of two EAHEC O104:H4 isolates, the closely related EAEC strain 55989 and the EHEC O157:H7 strain EDL933 to both cell lines. The microzins M and H47, which are produced by EcN, can be held responsible for a fraction of the observed anti-adhesive effect of EcN. The Microzins were also identified as the only substance that was influencing the growth of the pathogenic E. coli strains. One of the most important virulence factors of EHEC and EAHEC strains is Shiga toxin. In this study could be shown, that EcN is inhibiting the Shiga toxin production of the most common EHEC strains (big five: O157:H7, O26:H11, O103:H2, O111:H-, O145:H25) and two clinical isolates of EAHEC O104:H4 in the cell culture medium DMEM. Interesingly, the Shiga toxin 1 production of EHEC O103:H2 and O111:H- was not only reduced by EcN, but also the E. coli K-12 strain MG1655. In contrast, the Stx2 production of the serotypes O104:H4, O26:H11, O145:H25 was only blocked by EcN. The reduction of the Shiga toxin production in EAHEC O104:H4 TY3730 und TY3456, as well as EHEC O26:H11 was partly dependent on the microcin production of EcN. No influence of microzins on the Stx production of EHEC O157:H7 EDL933 and EHEC O145:H25 was detected. When using LB-medium instead of DMEM-medium, no influence of microzin on the Shiga toxin production of neither EAHEC TY3730 and TY3456, nor EHEC EDL933 could be shown. Experiments with SCEM-medium also resulted in an EcN-dose-dependent inhibition of Shiga toxin production of pathogenic E. coli strains in co-culture with EcN. In order to investigate the mechanism responsible for the observed effects, the EcN mutant EcN::luxS was used in co-culture experiments. However, the deletion of the quorum sensing molecule AI-2 in EcN::luxS had no influence on the Stx production. Using acetate in the experiments did not, in contrast to published results, lead to a reduction of Shiga toxin production. In addition, an influence of EcN on lysis of EHEC strains or the secretion of Shiga toxin could be ruled out. To study the Shiga toxin expression an assay with a bioluminescent C-P (chromosome-plasmid) reporter system was successfully established. Here, Shiga toxin expression could be monitored in real time with the strain background EHEC EDL933. Moreover, a reduction of Shiga toxin expression in co-culture with EcN could be detected. In further experiments could be shown, that EcN is not only reducing the Shiga toxin production in uninduced bacteria, but also in the Mitomycin C induced EAHEC O104:H4 strain TY3730. An important safety issue, in order to use EcN as a pharmaceutical against EHEC strains, is the resistance of EcN against Shiga toxin phages. The infection of bacteria was here investigated with phage plaque assay, stx-PCR, Stx-ELISA and K+-efflux assay. With these different methods could be successfully shown, that EcN is not infected by the tested Shiga toxin phages. A mutation in the phage receptor LamB could be a possible, but still unconfirmed, phage resistance mechanism of EcN. In summary, this study showed important antagonistic effects of EcN against pathogenic E. coli strains, which could be the foundation of new and desperately needed treatment options of EHEC infections.

No description available.

Includes bibliography.

The entrainment of microbes from agricultural land into overland flow during rainfall events is recognised as an important source of pathogenic microbes to surface water bodies and yet this transport process is poorly understood. In this study, a method has been developed to separate bacteria into the forms in which they have been postulated to exist in overland flow. Then Escherichia coli was used as a model organism to investigate the transported state of bacteria eroded from cowpats and their subsequent transport in overland flow. Simulated rainfall experiments were used to generate runoff direct from cowpats. Concentrations of E. coli in the runoff direct from cowpats were found to be directly proportional to the concentration in the cowpat, regardless of the age of the cowpat. It was also observed that E. coli were predominantly eroded from cowpats as individual cells. The interactions between E. coli and soil particles in overland flow were then examined in a small laboratory scale model system and showed that E. coli attached to large (>45 [mu]m) soil particles were transported significantly less than unattached cells. However, in the runoff from the model system, E. coli were found to be attached mainly to clay particles that were similar in size to the bacterial cells. Furthermore, the transport of E. coli through the model system appeared to follow the transport of a conservative chemical tracer implying that (a) the cells were being transported as a solute with the bulk of the water flow, and (b) that E. coli attached to small clay particles were as mobile in the overland flow as unattached cells. These observations imply that E. coli predominantly interact with small clay particles that are also being carried along in the overland flow. The transport of E. coli at a larger scale was then investigated using 5-metre long, 1-metre wide buffer strips operated under saturation excess conditions. In buffer strips using intact soils and existing pasture cover, E. coli removal was very poor (26 % removal) at the low flow rate of 2 L min⁻� with no removal observed at the higher flow rates of 6 and 20 L min⁻�. E. coli removal rates were increased to 41 % removal at 2 L min⁻� by cultivating the soils, with the removal rate again decreasing with increasing flow rate. E. coli in the overland flow from the buffer strips did not form into large flocs or attach to large soil particles, but were transported in small neutrally buoyant particles that remain entrained in the overland flow. Under saturation excess runoff conditions, E. coli in overland flow were not effectively removed by buffer strips as the small particles are transported either over the soil surface or, through large pores in the soil. This Thesis has shown that E. coli is transported in overland flow in small particle sizes that are difficult to trap or remove from overland flow thereby explaining the high fluxes of faecal bacteria observed in overland flow from agricultural land.

<p>Transmission of bacterial strains between patients is a serious problem in hospitals and with the increasing rate of antibiotic resistance the problem has farther escalated. Enterobacteriaceae produced extended-spectrum beta-lactamses (ESBLs), especially Escherichia coli (E-coli), are increasingly important nosocomial pathogens (7, 8). These bacteria are often multiple resistant and are responsible for many intestinal infections and urinary tract infections (2, 5). With the more frequent use of invasive devices in hospital care, these types of nosocomial infections have increased, particularly in seriously ill patients.</p><p>In order to diminish transmission of bacterial strains between patients and to study the epidemiology of these bacteria, it is of great importance to develop rapid and specific methods to be able to subtype on strain-level, i.e. to create a fingerprint of the isolates. The method may be based on phenotypic or genotypic characteristics of the microorganism. Any typing method must have high reproducibility and discrimination power to differentiate unrelated strains and also to demonstrate relationship of organisms deriving from the same source. In the present project, a Pulsed-Field Gel Electrophoresis (PFGE) assay for genotyping clinical E. coli isolates was used. PFGE can be used as a genotyping tool and is widely used to type bacteria and trace nosocomial infection. However, the method is time-consuming and relatively expensive in compare with other methods like PCR. In this study, a total of 93 strains were collected. The study was aimed to investigate the genotypes of the collected isolates and to identify and potential the outbreak strains.</p><p>The isolates investigated were genotypically diverse shown by a variety of PFGE banding patterns. However, clusters of closely related isolates involved in outbreaks were also identified.</p><p>In conclusion, when analyzing a large number of strains, a combination of a rapid phenotyping or genotyping method and a powerful genotyping method like PFGE would be an appropriate strategy for studying clonal relationship among isolates e.g. for detecting cross-transmission of nosocomial pathogens.</p>

Escherichia coli double-strand uracil-DNA glycosylase (Dug) was purified to apparent homogeneity from bacteria that were defective in uracil-DNA glycosylase (Ung). After cloning the dug gene, recombinant Dug was overexpressed, purified, and characterized with respect to activity, substrate specificity, product DNA binding, and mechanism of action. Purified Dug excised both uracil and ethenocytosine specifically from double-stranded DNA substrates. One distinctive characteristic of Dug was that the purified enzyme removed a near stoichiometric amount of uracil from DNA containing U/G mispairs. The observed lack of turnover was attributed to tight binding of Dug to the apyrimidinic-site (AP) contained in the DNA reaction product. Catalytic activity was stimulated in the presence of E. coli endonuclease IV that caused AP-site incision and dissociation of Dug. By using enzyme complementation experiments, Dug was shown to initiate uracil-initiated base excision repair (BER) in E. coli (ung) cell-free extracts. The relative rate of repair of uracil- and ethenocytosine-containing DNA in isogenic E. coli cells that were proficient or deficient in Ung and/or Dug was measured using a novel competition assay. Complete ethenocytosine-initiated BER displayed an absolute requirement for Dug and occurred at the same rate as uracil-initiated BER in the presence of both Ung and Dug. However, the rate of Dug-mediated ethenocytosine-DNA repair was 8-fold faster than that of uracil-DNA mediated by Dug. The distribution of BER patch sizes associated with both uracil- and ethenocytosine-containing DNA showed similar results. In both cases, DNA repair synthesis utilized predominantly a long patch BER mechanism involving the incorporation of 2-20 nucleotides. A previously unidentified "very long patch" mechanism of BER involving the incorporation of more than 200 nucleotides was identified and shown to be mediated by DNA polymerase I. The rate-limiting step associated with uracil-initiated BER was found to involve DNA ligase and the distribution of BER patch size was modulated by the ratio of DNA polymerase I and DNA ligase. The fidelity of DNA repair synthesis associated with complete uracil-DNA BER was measured using E. coli cell-free extracts that were proficient or deficient in Ung activity and determined to be 5.5 x 10������ and 19.7 x 10������, respectively. / Graduation date: 2003