Search

This thesis focuses on a special group of plant-beneficial pseudomonads with insecticidal activity. Bacteria belonging to the Pseudomonas fluorescens group are excellent root-colonizers with many plant beneficial effects. These bacteria can promote plant growth through the production of hormones and increase the availability of certain soil nutrients. Fluorescent Pseudomonas are also able to control the emergence of soilborne diseases mainly by the production of antimicrobial secondary metabolites and to induce systemic resistance in the plant host rendering it ready for the attack of pathogens or predators. Intriguingly, two species within the fluorescent pseudomonads, P. chlororaphis and P. protegens, possess the ability to colonize and kill Lepidopteran insects in addition to all these plant beneficial activities. In the last decade, several factors contributing to the insecticidal activity of these fascinating bacteria were identified, and insecticidal pseudomonads have emerged as promising candidates for the biological control of soil-derived insect pests for which no satisfactory control methods exist. The major aims of this thesis were to study the nature of Pseudomonas-insect interactions and its ecological significance, and to investigate, in a susceptible insect species, the pathogenicity process and the factors required at different phases during insect colonization and killing. In the first part of the thesis, we investigated whether the inability, of model strain P. protegens CHA0, to kill certain insect species, is due to its failure to persist in the animals. Based on feeding assays in combination with bacterial monitoring we showed that P. protegens CHA0 is highly lethal to larvae of Plutella xylostella (Lepidoptera) and Pieris brassicae (Lepidoptera) while being pathogenic, but less effective against larvae of Delia radicum (Diptera). P. protegens CHA0 had no effect on larvae of Otiorhynchus sulcatus (Coleoptera). However, P. protegens CHA0 persisted from larval to adult stage in all these species and was transmitted to a new plant host by D. radicum flies. These findings indicate that insecticidal pseudomonads can establish different relationships with insects ranging from exploiting insects as food source to using them as vectors for dispersal. A central part of the thesis concentrated on identifying the specific sets of genes a pseudomonad needs when colonizing a plant or an insect hosts, and to identify the specific role of individual factors during insect colonization and throughout disease progression. To this end, a large-scale transcriptomics dataset of P. protegens strain CHA0 was generated which includes data from the colonization of wheat roots, the gut of P. xylostella after oral uptake and the Galleria mellonella hemolymph after injection. The transcriptomic profiles strongly varied depending on the environment. Furthermore, we could associate specific factors to different hosts or different stages of insect infection. In addition, new traits contributing to insecticidal activity were identified, such as effector proteins (toxins) released by two-partner secretion systems (TPS). Their role during infection of P. xylostella was verified using a mutational approach. Altogether, this data allowed us to propose a comprehensive insect colonization and pathogenesis model for P. protegens CHA0. The ecological relevance of non-pathogenic interactions with insects described in the first part of the thesis, and the natural association of P. protegens and P. chlororaphis to insects, still remains unknown. To address these questions, we searched for insecticidal Pseudomonas in arthropods collected from a wheat field, a potato field and neighbouring grassland. We found that P. protegens and P. chlororaphis are naturally present in healthy insects and myriapods and isolated a set of new strains from arthropods, soil and roots. Although all strains, independently of their host of origin or their phylogenetic position, had oral activity against P. xylostella, the insecticidal capacities of different P. chlororaphis isolates were not as homogeneous as observed for P. protegens. Phylogenetically closely related P. chlororaphis strains differed in insect killing speed and efficiency. We hypothesize that the lower insecticidal activity observed for two Coleoptera isolates could be related to mutations in key insecticidal factors, such as the Fit toxin and TPSA proteins, that we discovered using a single nucleotide polymorphism analysis based on the whole genomes. These findings point towards an order-specificity or adaptation to certain insect hosts and show the ubiquitous nature of these special pseudomonads. This thesis substantially improves our knowledge about the pathogenesis of insect infecting Pseudomonas and the ecology of arthropod-Pseudomonas relationships. The novel information we gained is of great scientific, but also agricultural and environmental value, since it is highly important for the development of new biocontrol tools within the frame of a sustainable agriculture relying on environmentally friendly pest control methods.

Dioxygenases catalyze the dihydroxylation of aromatic compounds to the corresponding cis-dihydrodiols, which are interesting for organic synthesis but difficult to obtain by conventional chemical means. The topic of this thesis was the specific dihydroxylation of aromatic nitriles with chlorobenzene dioxygenase (CDO) from P. putida P51. We developed a recombinant biocatalyst Escherichia coli JM101 (pTEZ30) based on the alk regulatory system of P. putida Gpo1 for the efficient and stable expression of the CDO genes. Use of E. coli JM101 (pTEZ30) resulted in a significant increase in biocatalyst stability and volumetric productivity in the biotransformation of benzonitrile to cis-1,2-dihydroxy-3-cyanocyclohexa-3,5-diene (BNDD) in bioreactors. We further used E. coli JM101 (pTEZ30) for the regio- and stereospecific dihydroxylations of ortho-, meta-, and para- substituted benzonitriles, benzyl cyanide and cinnamonitrile. Biotransformations resulted in products with hydroxyl groups at 1,2 positions and with 42.9 - 97.1 % enantiomeric excess. Biotransformations of cinnamonitrile to trans-3-[(5S,6R)-5,6-dihydroxycyclohexa-1,3-dienyl]-acrylonitrile (CNDD) in bioreactors on 2-L and 30-L scales showed that the scale-up of cisdihydroxylation processes with E. coli JM101 (pTEZ30) is straightforward and once the efficient expression was achieved the processes became limited by the product toxicity. As a result of a screening program, a nitrilase from Rhodococcus sp. and AtNIT1 from Arabidopsis thaliana were found to hydrolyze the nitrile group of BNDD and CNDD to the corresponding acids. This enabled us to synthesize the novel chiral diols with an acid functional group. This work describes the development of a recombinant biocatalyst for efficient dihydroxylation processes, its use for the cis-dihydroxylation of different aromatic nitriles and its applications in bioreactors up to technical scales. It also discusses a possible combination of chlorobenzene dioxygenase and nitrilases, which provides a new route to a previously not described class of acids and chiral cis-diols. Dioxygenasen katalysieren die Dihydroxylierung organischer Substanzen zu ihren entsprechenden cis-Dihydrodiolen, welche eine vielversprechende Substanzklasse für organische Synthesen darstellt, aber zugleich auf konventionellem chemischem Wege schwer herstellbar ist. Das Thema dieser Doktorarbeit beschäftigte sich mit der spezifischen Dihydroxylierung aromatischer Nitrile mit Hilfe des Enzyms Chlorbenzoldioxygenase (CDO) aus Pseudomonas putida P51. Ausgehend vom alk Regulator System von P. putida GPo1 entwickelten wir den rekombinanten Stamm Escherichia coli JM101 (pTEZ30) für die effiziente und stabile Expression der CDO Gene. Durch die Verwendung von E. coli JM101 (pTEZ30) konnte dabei sowohl die Stabilität des Biokatalysators als auch die volumetrische Produktivität bei der Umsetzung von Benzonitril zu cis-1,2-Dihydroxy-3-cyanocyclohexa-3,5-dien (BNDD) im Bioreaktor gesteigert werden. Weiterhin konnten wir E. coli JM101 (pTEZ30) für die regio- und stereospezifische Dihydroxylierung von ortho-, meta- und para-substituierten Benzonitrilen sowie von Phenylacetonitril und Cinnamonitril nutzen. Die dabei gebildeteten Produkte waren spezifisch in 1,2-Position hydroxyliert und besassen Enantiomerenüberschüsse zwischen 42.9 - 97.1 %. Durch dieses effiziente Expressionssystem konnten wir weiterhin die Biotransformation von Cinnamonitril zu trans-3-[(5S,6R)-5,6-Dihydroxycyclohexa-1,3-dienyl]-acrylonitril (CNDD) im 2 und 20l Masstab zeigen und beweisen, dass eine Massstabsvergrösserung von cis-Dihydroxylierungen mit E. coli JM101 (pTEZ30) problemlos möglich ist und nun die Produkttoxizität limitierend wirkt. Als zusätzliches Ergebnis dieser Arbeit wurden in einem “Screening Programm” eine Nitrilase aus Rhodococcus sp. und die AtNIT1 Nitrilase aus Arabidopsis thaliana gefunden, die die Nitrilgruppe der erhaltenen Dihydrodiole BNDD und CNDD zur entsprechenden Säure hydrolysieren konnten, was uns befähigte, die neuen chiralen Diole mit einer zusätzlichen funktionellen Säure-Gruppe zu synthetisieren. Damit beschreibt die vorliegende Arbeit die Entwicklung eines rekombinanten Biokatalysators für effiziente Dihydroxylierungsprozesse, deren Anwendung mit verschiedenen aromatischen Nitrilen und deren Einsatz in Bioreaktoren bis zum technischen Massstab. Weiterhin wird die Kombination von Chlorbenzoldioxygenase und Nitrilasen vorgestellt, wodurch eine neue Syntheseroute für die Klasse der chiralen cis-Dihydroxysäuren ermöglicht wird.