Refine
Year of publication
Document Type
- Doctoral Thesis (30)
Has Fulltext
- yes (30)
Is part of the Bibliography
- no (30)
Keywords
- Xenorhabdus (2)
- 3-alkylphenols (1)
- 6-methylsalicylic acid synthase (1)
- Fabclavine (1)
- Fatty acid synthases (1)
- Inthraszentin (1)
- Natural Products (1)
- Natural products (1)
- Naturstoffe (1)
- Photorhabdus (1)
Institute
Natural products are valuable sources for biologically active compounds, which can be utilized as pharmaceuticals. Thereby, the synthesis is based purely on biosynthetic grounds often conducted by so-called megaenzymes. One major biosynthetic pathway is the acetate pathway including polyketide and fatty acid synthesis, which encompass one of the largest classes of chemically diverse natural products. These have medicinal relevance due to their antibacterial, antifungal, anthelmintic, immunosuppressive and antitumor properties.
Due to the high structural and functional similarity between polyketide synthases and type I animal fatty acid synthases (FASs), FAS can serve as a paradigm for the whole class of multifunctional enzymes. To fully exploit the biosynthetic potential of FASs, a good access to the enzyme is of essential importance. In this regard, Escherichia coli remains an unchallenged heterologous host due to low culturing costs, particularly fast mutagenesis cycles and relatively easy handling. Surprisingly, no sufficient expression strategy for an animal FAS in E. coli has yet been reported, as it turned out that the only approach was not reproducible.
We commenced our analysis with searching for an appropriate FAS homolog that fulfills our requirements of high protein quality, sufficient yield and ensured functionality. After extensive screening of different variants, culturing conditions and co-expression strategies, we identified the murine FAS (mFAS) as our protein of choice. The established purification strategy using tags at both termini led to a reproducible and sufficient access to the protein in excellent quality. The enzyme was further biochemically characterized including an enzyme kinetic investigation of fatty acid synthesis and an examination whether different acyl-CoA substrates can serve as priming units. This adds mFAS to our repertoire of manageable megaenzymes paving the way to exploit the catalytic efficiency in regards of microbial custom-compound synthesis.
With a strong focus on deepening our understanding of the working mode of such megaenzymes, rather than analyzing respective biosynthetic products, we have addressed the question whether mFAS itself can be engineered towards PKSs or whether properties of mFAS can be exploited to engineer PKSs. This approach was conducted on three levels of complexity from function of individual domains via organization of domains to form modules to the interplay of two modules in bimodular constructs.
Fatty acid synthesis begins with the loading of acyl moieties onto the FAS, which is conducted by a domain called malonyl-/acetyltransferase (MAT). This domain was in-depth characterized due to its important role of choosing the substrates that are built in the final compound. Our analysis comprised structural and functional aspects providing crystal structures of two different acyl-bound states and kinetic parameters for the hydrolysis and transacylation reaction using twelve exemplary CoA-esters. For this purpose, we have successfully established a continuous fluorometric assay using the α-ketoglutarate dehydrogenase as a coupled enzyme, which converts the liberated coenzyme A into Nicotinamide adenine dinucleotide. These data revealed an extensive substrate ambiguity of the MAT domain, which had not been reported to that extent before. Further, we could demonstrate that the fold fulfills both criteria for the evolvability of an enzyme by expressing MAT in different structural arrangements (robustness) and by altering the substrate ambiguity within a mutagenesis study (plasticity). Taken these aspects together, we are persuaded that the MAT domain can serve as a versatile tool for PKSs engineering in potential FAS/PKS hybrid systems.
On the higher level of complexity, we investigated the architectural variability of the mFAS fold, which constitutes a fundamental basis for a broader biosynthetic application. We could rebuild all four module types occurring in typical modular PKSs confirming a high degree of modularity within the fold. Not only structural, but also functional integrity of these modules was validated by using triacetic acid lactone formation and ketoreductase activity. Especially the latter analysis, made it possible to quantify effects of the engineering within the processing part by respective enzyme kinetic parameters. Expanding our focus beyond a singular module, we have utilized the mFAS fold for designing up to 380 kDa large bimodular constructs. In this approach, a loading didomain was attached N-terminally containing an additional MAT and acyl carrier protein (ACP) domain. Two constructs could be expressed and purified in excellent quality to investigate the influence of an altered overall architecture on fatty acid synthesis. By comparison with appropriate controls, a functional effect of the additional loading module could indeed be proven in the bimodular systems. Those constructs allow a comprehensive analysis of the underlying molecular mechanism in the future and serve as a potential model system to study the transition from iterative to vectorial polyketide synthesis in vitro.
1. Das Genom von A. woodii konnte sequenziert und annotiert werden. Der Organismus besitzt ein Chromosom von 4050521 Bp und keine Plasmide. Es sind 3495 ORFs kodiert. 2. Die Gene, die die Enzyme des Wood-Ljungdahl-Weges kodieren, konnten identifiziert werden. Sie sind hauptsächlich in drei Clustern organisiert, wobei für Cluster II gezeigt werden konnte, dass es ein Operon bildet und dort ungewöhnlicherweise ein RnfC-ähnliches Protein kodiert ist. 3. Gene für Proteine der Hexose-Verwertung konnten ebenfalls identifiziert werden. A. woodii besitzt sowohl PTS-Systeme als auch einen Na+/Zucker-Symporter zur Aufnahme von Hexosen. Die Enzyme der Glykolyse sind vollständig im Genom vorhanden und liegen im gesamten Genom verstreut vor. 4. Neben den Genen für die bereits charakterisierte Hydrogenase existieren im Genom weitere Gene, die potentielle Hydrogenasen oder Untereinheiten dieser kodieren. 5. Lange wurde für Methyltransferasen in A. woodii vermutet, dass es sich um energiekonservierende Enzyme handelt. Die Genomsequenz zeigte, dass das Genom Gene für 20 Methyltransferasen 1, 10 Methyltransferasen 2 und 22 Corrinoid-Proteine enthält. Die Methyltransferase und das Corrinoid-Protein des Wood-Ljungdahl-Weges konnten identifiziert werden. Allerdings konnte für keines der korrespondierenden Proteine eine Membranständigkeit vorhergesagt werden, was eine Beteiligung der Methyltransferasen an der Energiekonservierung ausschließt. Die Vielzahl der Methyltransferasen passt aber zu der Vielzahl von methylierten Verbindungen, die der Organismus verstoffwechseln kann. 6. Neben den gut charakterisierten etf-Genen aus dem car-Operon, das bei der Caffeat-Reduktion eine wichtige Rolle spielt, gibt es ein weiteres etf-Paar, welches mit den Genen für eine Laktat-Dehydrogenase und eine Laktat-Permease kolokalisiert ist. Welche Rolle die Proteine spielen bleibt noch aufzuklären. 7. Außer den Genen für die gut charakterisierte F1F0-ATP-Synthase finden sich Gene für eine V-Typ ATPase. Diese Gene bilden ein Operon. Desweiteren konnte gezeigt werden, dass die Untereinheit VatA auch produziert wird. Die physiologische Rolle konnte allerdings noch nicht geklärt werden. 8. Basierend auf den genomischen Daten konnte ein Modell des Flagellums erstellt werden. Desweiteren wurde eine Vielzahl von Genen für chemotaktische Proteine identifiziert. Zur Verarbeitung von Umweltsignalen besitzt A. woodii Komponenten des Che-Systems, die zum einen aus E. coli und zum anderen aus B. subtilis bekannt sind. 9. In Proteomanalysen konnte festgestellt werden, dass die Enzyme des Wood- Ljungdahl-Weges beim Wachstum auf H2 + CO2 im Vergleich zum Wachstum auf Fruktose induziert werden, die Enzyme der Glykolyse werden dagegen reprimiert. Desweiteren ist die Hydrogenase (HydAB) auf H2 + CO2 induziert. Das am stärksten induzierte Protein ist eine Alanin-Dehydrogenase, deren Rolle im Stoffwechsel unbekannt ist. 10. Die Untersuchung des genomischen Kontextes der für die Na+-translozierende Ferredoxin:NAD+-Oxidoreduktase (Fno/Rnf) kodierenden Gene rnfCDGEAB ergab keine weiteren Gene, die mit Rnf in Verbindung stehen. Experimentelle Befunde zeigen, dass die Gene rnfCDGEAB ein Operon bilden. 11. Nach der Generierung von Antikörpern gegen die Untereinheiten des Rnf-Komplexes, die große lösliche Anteile besitzen, konnte nachgewiesen werden, dass RnfB, C und G in der Membran lokalisiert sind. Desweiteren wurde nachgewiesen, dass deren Produktion unabhängig von der An- oder Abwesenheit von Caffeat und den getesteten C-Quellen ist. 12. RnfG konnte in E. coli überproduziert und anschließend gereinigt werden, allerdings fehlte der vorhergesagte, kovalent gebundene Flavin-Cofaktor. 13. RnfC konnte ebenfalls in E. coli überproduziert und anschließend gereinigt werden. Nach Rekonstitution mit Eisen und Schwefel konnte ein Fe-Gehalt von 8 nmol/ nmol Protein und ein Schwefel-Gehalt von 5 nmol/nmol Protein bestimmt werden. Die im UV/Vis-Spektrum sichtbaren Maxima wiesen auf die Anwesenheit von FeS-Zentren hin. EPR-Analysen deuten darauf hin, dass die FeS-Zentren nur unvollständig assembliert sind. 14. Im Genom von A. woodii ist ein Cluster von Genen, das Proteine zur Umsetzung von 1,2-Propandiol kodiert, zu finden. Elektronenmikroskopisch konnte nachgewiesen werden, dass der Organismus in Gegenwart von 1,2-Propandiol Mikrokompartimente bildet. 15. In Zellsuspensionsversuchen konnte nachgewiesen werden, dass 1,2-Propandiol nicht zu Propionat und Acetat, sondern zu 1-Propanol und Propionat über das Intermediat Propionaldehyd umgesetzt wird. 16. Rohextrakte 1,2-Propandiol-gezogener Zellen katalysierten die Reduktion von NAD+ mit Propionaldehyd als Reduktant. Die Reaktion benötigte CoA, NAD+ (Km 0,35 mM) und Propionaldehyd (Km 1,3 mM). Das Temperaturoptimum betrug 30°C und das pH-Optimum lag zwischen pH 8 und 10. 17. Ein Antikörper gegen die Propionaldehyd-Dehydrogenase (PduP) aus S. enterica reagierte mit einem ca. 50 kDa-Protein 1,2-Propandiol-gezogener Zellen. Dies zeigt, dass PduP aus A. woodii und PduP aus S. enterica immunologisch verwandt sind. Western-Blot-Analysen zeigten, dass PduP nur in 1,2-Propandiol-, 2,3-Butandioloder Ethylenglykol-gezogenen Zellen nachweisbar war, aber nicht in Zellen die auf Fruktose, Ethanol oder H2 + CO2 gezogen waren. 18. Die Aktivität der Propionaldehyd-Dehydrogenase war in Zellen gezogen auf 1,2-Propandiol am höchsten. Nach Wachstum auf Fruktose oder H2 + CO2 war die Aktivität sehr niedrig. Genau gegensätzlich verhielten sich die Aktivitäten der Formiat-Dehydrogenase, einem Enzym des Wood-Ljungdahl-Weges, der ATPHydrolyse und des Rnf-Komplexes. 19. In Gegenwart von Caffeat und 1,2-Propandiol konnte A. woodii nicht wachsen. Das Wachstum auf 2,3-Butandiol oder Ethylenglykol in Gegenwart von Caffeat war möglich.
As fossil resources are diminishing, environmental concerns arise and chemical synthesis often involves expensive catalysts or extensive extraction procedures, the demand for production of industrially relevant compounds from renewable resources increases. In this context, engineering microorganisms for production of specialty chemicals, such as 3-alkylphenols, presents an attractive, environmental-friendly approach. 3-alkylphenols have various applications: due to their antiseptic and stabilizing properties many 3-alkylphenols, including 3-methylphenol (3-MP), are utilized as additives in disinfectant reagents and biological products, while they can be also implemented as platform chemicals for production of lubricating oil additives or flavors. Some 3-akylphenols have potential for transmission control of the disease sleeping sickness that is transmitted by tsetse flies in sub-saharan Africa, since 3-ethylphenol (3-EP) and 3-propylphenol (3-PP) and to a lesser degree 3-MP were found to attract tsetse flies and improved catch rates in impregnated tsetse fly traps. Microbial fermentation of 3-alkylphenols would provide a simple and inexpensive way for local communities in Africa to produce these compounds and prepare their own tsetse fly traps.
Some molds synthesize 3-MP as an intermediate during biosynthesis of the mycotoxin patulin. However, the heterologous host Saccharomyces cerevisiae has advantageous traits for industrial application, since it is well characterized, robust, simple to handle and easily genetically accessible. In this thesis, genetical engineering approaches were utilized to establish the yeast S. cerevisiae for biotechnological production of 3-alkylphenols. As a proof of concept, the iterative polyketide synthase from Penicillium patulum, 6-methylsalicylic acid synthase (MSAS), and 6-methylsalicylic acid (6-MSA) decarboxylase PatG from Aspergillus clavatus were heterologously expressed in S. cerevisiae resulting in the first reported de novo biosynthesis of 3-MP via 6-MSA in yeast from sugars (Hitschler & Boles, 2019). It was shown that codon-optimization and genomic integration of heterologous genes, high initial cell densities and a balanced expression of PatG were beneficial for heterologous production of up to 589 mg/L 3-MP in S. cerevisiae. However, toxicity of 3-MP limited higher product accumulation.
Different in vivo detoxification strategies were implemented to face this bottleneck. Growth tests revealed that 3-methylanisole (3-MA) is less toxic to the yeast cells than 3-MP. Expression of an orcinol-O-methyltransferase from chinese rose hybrids (OOMT2) was combined with in situ extraction converting the toxic 3-MP product into the volatile 3-MA and accumulating up to 211 mg/L 3-MA in the dodecane phase. Alternatively, up to 533 mg/L 3-MP glucoside were synthesized by expression of a UDP-glycosyltransferase (UGT72B27) from Vitis vinifera in the 3-MP producing strain, revealing saccharose as beneficial carbon source and ethanol growth phase as essential for high 3-MP production, although 3-MP conversions were not yet complete. Both detoxification strategies allowed circumvention of the toxicity imposed limited product accumulation. This was demonstrated when both detoxification strategies were combined with redirection of the carbon flux through deletion of phosphoglucose isomerase gene PGI1 and feeding a mixture of fructose and glucose leading to majorly improved product formation, with up to 899 mg/L 3-MA/3-MP and 873 mg/L 3-MP/3-MP glucoside, compared to less than 313 mg/L product titers in the wild type controls (Hitschler & Boles, 2020).
For provision of the tsetse fly attractants 3-EP from propionyl-CoA and 3-PP from butyryl-CoA, the substrate promiscuities of MSAS and PatG were exploited. However, slower formation rates with the alternative substrates propionyl-CoA and butyryl-CoA suggested that competing formation of 6-MSA from the preferred priming unit acetyl-CoA was dominating in vivo. Indeed, 3-EP or 3-PP formation was not observed in 3-MP producing yeast strains. Assuming that intracellular levels of propionyl-CoA and butyryl-CoA were limiting 3-EP and 3-PP formation, different strategies were implemented to raise the supply of these alternative priming units and successfully compete with acetyl-CoA for MSAS priming.
Supplementation of propionate increased propionyl-CoA levels by endogenous pathways sufficiently to enable 3-EP formation in yeast mediated by MSAS and PatG. Deletion of the 2-methylcitrate synthases CIT2 and CIT3 revealed that degradation of propionyl-CoA was not limiting 3-EP formation at this stage. In order to raise propionyl-CoA levels further, a heterologous propionyl-CoA synthase (PrpE) was expressed in the 3-MP producing yeast strain leading to up to 12.5 mg/L 3-EP with propionate feeding and blockage of degradation. Moreover, PrpE enabled also 3-EP formation without propionate supplementation suggesting that an endogenous supply of propionate existed that was reactivated by PrpE. As threonine or 2-ketobutyrate feeding increased 3-EP titers in combination with PrpE, this indicated that threonine degradation via 2-ketobutyrate was responsible for the endogenous propionate supply. Moreover, expression of branched-chain ketoacid dehydrogenase complex from Pseudomonas putida combined with PrpE provided propionyl-CoA from endogenous 2-ketobutyrate and raised 3-EP titers up to 5.9 mg/L compared to 2.8 mg/L with only PrpE indicating a potential route for optimization of 3-EP titers independent of propionate or threonine feeding.
For 3-PP production from butyryl-CoA, a heterologous ‘reverse ß-oxidation’ pathway was introduced in the 3-MP producing yeast strain providing sufficient butyryl-CoA for biosynthesis of up to 2 mg/L 3-PP. Degradation of the precursor via ß-oxidation was slightly limiting, since deletion of fatty acyl-CoA oxidase POX1 increased 3-PP titers slightly to 2.6 mg/L.
As the concentrations of 3-alkylphenols are close to the concentrations implemented in tsetse fly traps, the engineered yeast strains have the potential for simple and inexpensive on-site production of 3-alkylphenols as tsetse fly attractants by local rural communities in Africa. In spite of this success, 3-MP remained the main product in the developed yeast strains. Since 3-EP and 3-PP are more efficient tsetse fly attractants, a shift in substrate specificities of MSAS and PatG is desirable for a more favorable 3-EP/3-MP and 3-PP/3-MP product ratio regarding tsetse fly attraction. During rational engineering of MSAS, the MSASQ625A/I752V mutant showed a beneficial shift of product ratios with up to 11 mg/L 3-EP/63 mg/L 3-MP and 4.5 mg/L 3-PP/116 mg/L 3-MP, compared to a higher proportion of 3-MP with up to 343 mg/L, 11 mg/L 3-EP and 1.5 mg/L 3-PP in the wild type controls. Further engineering of MSAS and PatG might majorly improve production of 3-EP and 3-PP.
In summary, this thesis successfully established the yeast S. cerevisiae as cell factory for production of different 3-alkylphenols optimizing expression of the heterologous production pathway, elucidating means to detoxify products and establishing different approaches to increase intracellular levels of acyl-CoA precursors. The engineered yeast strains can be potentially implemented for simple and inexpensive fermentation of tsetse fly attractants in Africa.
The growing number of infections with multi-resistant bacteria or the current COVID-19 pandemic put compounds with therapeutic properties into the public focus. Non-ribosomal peptides (NRPs) are natural products that are already marketed as antibiotics, cytotoxic agents or immunosuppressants. Their biological activities rely on the structural diversity including non-proteinogenic amino acids (AAs), heterocycles or modifications like methylation or acylation.
The biosynthesis of NRPs is carried out by non-ribosomal peptide synthetases (NRPSs). These multifunctional megaenzymes show a modular architecture like in an assembly-line. Each module is thereby responsible for the incorporation and modification of one AA and therefore contains different catalytic domains. The adenylation (A) domain recognizes and activates its specific substrate in an ATP-dependent manner which is transferred to a 4’-phosphopantetheine cofactor post-translationally attached to the thiolation (T) domain. Peptide bond formation between two T domain bound substrates catalysed by the condensation (C) domain transfers the growing peptide chain to the following module. Such a C-A-T module can be extended with optional domains to integrate structural diversity and a terminal thioesterase (TE) domain usually releases the peptide via hydrolysis or intramolecular attack of nucleophiles. Inspired by the modular architecture, NRPS engineering deals with the modification of NRPs in order to increase biological activities, circumvent bacterial resistances or create de novo peptides. This can be achieved by mutasynthesis or modification of the substrate binding pocket as well as single and multiple domain substitution. However, the few successful approaches led to impaired enzymes and did not establish a general applicable guideline. In the first publication as part of this work, the development of such a guideline comprising three rules is addressed. First, the A-T-C tridomain named exchange unit (XU) is seen as a catalytic unit instead of a module. When using them as building blocks, the C domain’s specificity for the AA of the following XU has to be considered as second rule. Third, a conserved WNATE motif within the C-A linker depicts the fusion point of the XUs. Upon heterologous expression of the cloned plasmids in E. coli and high performance liquid chromatography coupled mass spectrometry-based analysis of the extracts, the ambactin-producing NRPS from Xenorhabdus was reprogrammed with one and two XUs. This only leads to a moderate loss of production titre or an even higher one when the AA configuration was changed by introducing a dual condensation/epimerization (C/E) domain. The pentamodular GameXPeptide-producing NRPS was reconstructed using up to five XUs of four different NRPSs and even completely de novo synthetases were created. The second publication describes the exchange unit condensation domain (XUC) concept and relies on a fusion point between the two subdomains (N-terminal CDsub and C-terminal CAsub) of the C domain’s V-shaped pseudodimeric structure which generates A-T didomains with flanking CAsub and CDsub. These hybrid C domain-forming building blocks depict an improvement to the XU concept by avoiding the drawback of C domain specificity. This allows a more flexible NRPS engineering that can e.g. enable peptide library design. Furthermore, beside a combination of both concepts within one NRPS and a transfer to Bacillus NRPSs, the use of XUC with relaxed A domain specificity allowed further peptide modifications by introducing non-natural AAs. The third publication deals with aldehyde and alcohol-generating reductase (R) domains which depict an alternative for peptide release in NRPSs. A promoter exchange in X. indica identified a pyrazine-producing NRPS with a minimal architecture of an A, T and R domain and was therefore termed ATRed. R domains were additionally used in engineered NRPSs to produce pyrazinones and derivatives thereof by XU substitution although most constructs failed to show production. Beyond that, an R domain has been shown to replace a TE domain in wild type synthetases leading to slightly modified NRPs and the postulated biosynthesis was incidentally revised. Furthermore, an NRPS with terminal R domain was engineered to produce a free peptide aldehyde, which are known to be potent proteasome inhibitors. For the above mentioned ATReds, the presence of up to three coding regions was further identified in 20 different Xenorhabdus strains but only six of them were verified to produce pyrazines. All ATReds share variable sequence similarities among each other and were subsequently divided into three subtypes. One subtype is supposed to perform the pyrazine biosynthesis via a non-canonical catalytic triad.
Nematophilic bacteria as a source of novel macrocyclised antimicrobial non-ribosomal peptides
(2020)
A solution to ineffective clinical antimicrobials is the discovery of new ones from under-explored sources such as macrocyclic non-ribosomal peptides (NRP) from nematophilic bacteria. In this dissertation an antimicrobial discovery process –from soil sample to inhibitory peptide– is demonstrated through investigations on six nematophilic bacteria: Xenorhabdus griffiniae XN45, X. griffiniae VH1, Xenorhabdus sp. nov. BG5, Xenorhabdus sp. nov. BMMCB, X. ishibashii and Photorhabdus temperata. To demonstrate the first step of bacterium isolation and species delineation, endosymbionts were isolated from Steinernema sp. strains BG5 and VH1 that were isolated directly from soil samples in Western Kenya. After genome sequencing and assembly of novel Xenorhabdus isolates VH1 and BG5, species delineation was done via three overall genome relatedness indices. VH1 was identified as X. griffiniae VH1, BG5 as Xenorhabdus sp. nov. BG5 and X. griffiniae BMMCB was emended to Xenorhabdus sp. nov. BMMCB. The nematode host of X. griffiniae XN45, Steinernema sp. scarpo was highlighted as a putative novel species. To demonstrate the second step of genome mining and macrocyclic non-ribosomal peptide structure elucidation, chemosynthesis and biosynthesis, the non-ribosomal peptide whose production is encoded by the ishA-B genes in X. ishibashii was investigated. Through a combination of refactoring the ishA-B operon by a promoter exchange mechanism, isotope labelling experiments, high resolution tandem mass spectrometry analysis, bioinformatic protein domain analysis and chemoinformatic comparisons of actual to hypothetical mass spectrometry spectra, the structures of Ishipeptides were elucidated and confirmed by chemical synthesis. Ishipeptide A was a branch cyclic depsidodecapeptide macrocyclised via an ester bond between serine and the terminal glutamate. It chemosynthesis route was via a late stage macrolactamation and linearised Ishipeptide B was synthesised via solid phase iterative synthesis. Ishipeptides were not N-terminally acylated despite being biosynthesised from the IshA protein that had a C-starter domain. It was highlighted that more than restoration of the histidine active site of this domain is required to restore N-terminal acylation activity.
To demonstrate the final step of determination of antimicrobial activity, minimum inhibitory concentrations of Ishipeptides and Photoditritide from Photorhabdus temperata against fungi and bacteria were determined. None were antifungal while only the macrocyclic compounds were inhibitory, with Ishipeptide A inhibitory to Gram-positive bacteria at 37 µM. The cationic Photoditritide, a cyclic hexapeptide macrocyclised via a lactam bond between homoarginine and tryptophan, was 12 times more inhibitory (3.0 µM), even more effective than a current clinical compound, Ampicillin (4.2 µM). For both, macrocyclisation was hypothesised to contribute to antimicrobial activity. Ultimately, this dissertation demonstrated not only nematophilic bacteria as a source of novel macrocyclic antimicrobial non-ribosomal peptides but also a process of antimicrobial discovery–from soil sample to inhibitory peptide– from these useful bacteria genera. This is significant for the fight against antimicrobial resistance.
Photorhabdus and Xenorhabdus bacteria live in a highly specific symbiosis with nematodes that belong to the genus of Heterorhabditis and Steinernema, respectively. These cruiser type nematodes actively search for soil-dwelling insects and infect them via natural openings. Inside of the insect, the bacteria are released into the hemocoel where they start producing an array of secondary metabolites to bypass the insect immune system and kill the prey within 48 hours. Many of those natural products possess bioactivities against other bacteria, fungi, protozoa or insects, which makes them interesting candidates for pharmaceutical applications. Even though advanced molecular biological methods in combination with bioinformatics tools can now be used to predict biosynthetic gene clusters (BGCs) and their products, there are still many BGCs with unknown products. Even for the plethora of natural products that were successfully identified in the last couple of years, the exact ecological function often remains elusive, as laboratory conditions can vary considerably from the natural environment of the bacteria. Knowledge about the natural conditions that stimulate, or repress production of certain natural products and their underlying regulatory mechanisms yield new approaches for natural product research and enables possibilities for selective manipulations of the regulatory cascades.
The overarching goal of this work was to examine the regulatory networks in Photorhabdus and Xenorhabdus strains. The first part of this work focused on the Hfq-dependent regulation of specialized metabolite production. In those genera, the RNA chaperone, Hfq, represses expression of hexA, which encodes for a global transcriptional regulator that acts as the master repressor for SM production. Multiple global approaches were used to identify the sRNA ArcZ, which targets a specific region in the 5’-untranslated region of the hexA mRNA and ultimately guides Hfq in order to repress its expression. It was shown that a deletion of arcZ led to a drastic reduction of SM production in Photorhabdus and Xenorhabdus, consistent with the phenotype of their respective hfq deletion mutants. Transcriptomic profiling revealed far-reaching effects on the transcriptome, with up to 735 coding sequences significantly affected in the arcZ deletion strain. Finally, it was shown that the resulting chemical background, devoid of SMs, in combination with targeted promotor exchange can be used to exclusively overproduce a desired natural product, representing an alternative route of genetic manipulation.
The second part of this work focused on the influence and identification of insect related compounds that affect SM production in P. laumondii, X. szentirmaii and X. nematophila. Insect homogenate was generated from G. mellonella larvae, a model host for these bacteria. Supplementation of the cultivation medium with homogenate induced considerable shifts in the SM profiles of those bacteria. A global effect on the transcriptional output was determined by transcriptomic profiling. The core response to the simulation of an insect environment consisted of ten CDS, eight of which are involved in the degradation of fatty acids or the import of maltose and maltodextrin into the cells. Two abundant components in the insect homogenate, trehalose and putrescin, were added to the cultivation medium of those strains and subsequent HPLC-MS analysis revealed a direct correlation of their concentration in the medium and the production titres of certain SMs. These results indicated that the bacteria sense the insect environment via different insect specific components in order to initiate a metabolic adjustment, which is probably required for adaptation to the insect host.
The last part of this work examined the influence of other, so far not directly related genes on SM production, based on the isolation of P. laumondii transposon-insertion mutants with clear phenotypic alterations. Re-sequencing and SM profiling of the mutant strains revealed that a transposon-insertion in the gene encoding for a putative DNA-adenine methyltransferase affected SM production. The phenotype was confirmed by deleting this gene. Based on Single-Molecule Real-Time sequencing, the complete methylome of the WT, deletion- and complementation mutant were analysed (experimental work performed by Sacha J. Pidot, Melbourne, Australia). No obvious alterations were detected in the methylation patterns of the strains, indicating that the dam gene product does not methylate the adenine in GATC-motifs, as it was described in literature for E. coli. This data raises the question what the function of the putative DNA-adenine methyltransferase is in P. laumondii and how it can influence the secondary metabolism. Even though there is currently no clear evidence, the potential role of epigenetic gene regulation mechanisms should be considered in further work.
The application of natural products (NPs) as drugs and lead compounds has greatly improved human health over the past few decades. Despite their success, we still need to find new NPs that can be used as drugs to combat increasing drug resistance via new modes of action and to develop safer treatments with less side effects.
Entomopathogenic bacteria of Xenorhabdus and Photorhabdus that live in mutualistic symbiosis with nematodes are considered as promising producers of NPs, since more than 6.5% of their genomes are assigned to biosynthetic gene clusters (BGCs) responsible for production of secondary metabolites. The investigation on NPs from Xenorhabdus and Photorhabdus can not only provide new compounds for drug discovery but also help to understand the biochemical basis involved in mutualistic and pathogenic symbiosis of bacteria, nematode host and insect prey.
Nonribosomal peptides (NRPs) are a large class of NPs that are mainly found in bacteria and fungi. They are biosynthesized by nonribosomal peptide synthetases (NRPSs) and display diverse functions, representing more than 20 clinically used drugs. Although a large number of NRPs have been identified in Xenorhabdus and Photorhabdus, the advanced genome sequencing and bioinformatic analysis indicate that these bacteria still have many unknown NRPS-encoding gene clusters for NRP production that are worth to explore. Therefore, this thesis focuses on the discovery, biosynthesis, structure identification, and biological functions of new NRPs from Xenorhabdus and Photorhabdus.
The first publication describes the isolation and structure elucidation of seven new rhabdopeptide/xenortide-like peptides (RXPs) from X. innexi, incorporating putrescine or ammonia as the C-terminal amines. Bioactivity testing of these RXPs revealed potent antiprotozoal activity against the causative agents of sleeping sickness (Trypanosoma brucei rhodesiense) and malaria (Plasmodium falciparum), making them the most active RXP derivatives known to date. Biosynthetically, the initial NRPS module InxA might act iteratively with a flexible methyltransferase activity to catalyze the incorporation of the first five or six N-methylvaline/valine to these peptides.
The second publication focuses on the structure elucidation of seven unusual methionine-containing RXPs that were found as minor products in E. coli carrying the BGC kj12ABC from Xenorhabdus KJ12.1. To confirm the proposed structures from detailed HPLC-MS analysis, a solid-phase peptide synthesis (SPPS) method was developed for the synthesis of these partially methylated RXPs. These RXPs also exhibited good effects against T. brucei rhodesiense and P. falciparum, suggesting RXPs might play a role in protecting insect cadaver from soil-living protozoa to support the symbiosis with nematodes.
The third publication presents the identification of a new peptide library, named photohexapeptide library, which occurred after the biosynthetic gene phpS was activated in P. asymbiotica PB68.1 via promoter exchange. The chemical diversity of the photohexapeptides results from unusual promiscuous specificity of five out of six adenylation (A) domains being an excellent example of how to create compound libraries in nature. Furthermore, photohexapeptides enrich the family of the rare linear D-/L-peptide NPs.
The fourth publication concentrates on the structure elucidation of a new cyclohexapeptide, termed photoditritide, which was produced by P. temperata Meg1 after the biosynthetic gene pdtS was activated via promoter exchange. Photoditritide so far is the only example of a peptide from entomopathogenic bacteria that contains the uncommon amino acid homoarginine. The potent antimicrobial activity of photoditritide against Micrococcus luteus implies that photoditritide can protect the insect cadaver from food competitor bacteria in the complex life cycle of nematode and bacteria.
The last publication reports a new family of cyclic lipopeptides (CLPs), named phototemtides, which were obtained after the BGC pttABC from P. temperata Meg1 was heterologously expressed in E. coli. The gene pttA encodes an MbtH protein that was required for the biosynthesis of phototemtides in E. coli. To determine the absolute configurations of the hydroxy fatty acids, a total synthesis of the major compound phototemtide A was performed. Although the antimalarial activity of phototemtide A is only weak, it might be a starting point towards a selective P. falciparum compound, as it shows no activity against any other tested organisms.
This work deals with the characterization of three different type II polyketide synthase systems (PKS II) from the Gram-negative bacteria Xenorhabdus and Photorhabdus.
Particular attention was paid to a biochemically underexplored class of aryl polyene (APE) pigments. Bioinformatic analysis of enzymes involved in the biosynthesis and the in vitro reconstruction proved that the synthesis of APEs involves an unusual fatty acid-like elongation mechanism. Furthermore, the discovery of unexpected protein-protein interactions provided new insights into the multienzyme complex formation of this unusual PKS II system. Through collaboration with the groups from Prof. Michael Groll and junior Prof. Nina Morgner, two protein complexes were structurally solved and several native protein multimerization events were identified and allowed us to suggest a possible protein-interaction network. The results are summarized in publication ‘An Uncommon Type II PKS Catalyzes Biosynthesis of Aryl Polyene Pigments’ (first author; J. Am. Chem. Soc.).
In addition to in vitro-analysis, in vivo-studies were used to investigate the APE compound produced by X. doucetiae in more detail. The activation of the silent biosynthetic gene cluster (BGC) led to the detection of the APE compound in the homologous host. Further combination of homologous expression and targeted deletions of the APE BGC revealed an APE-lipid-like structure. MS-based analyses and purification of intermediates allowed us to deduce structural building blocks of the APE-lipid, which is composed of an APE structural core, a glucosamine residue and an unusual long-chain fatty acid with unusual conjugated double bonds and a phosphoethanolamine head group. In combination with the above stated in vitro-data, we assumed a plausible biosynthetic mechanism of the APE-lipid. The results are summarized in the section ‘Additional Results: Tracing the Full-length APE’.
The biosynthesis of isopropylstilbene (IPS) has already been well-studied by the Bode laboratory and the group of Prof. Ikuro Abe. Studies with Photorhabdus laumondii TT01 by the Bode group revealed the distributed locations and functions of the genes involved in biosynthesis, which originate from two pathways. Particularly, the Bode group first demonstrated that an unusual ketosynthase/cyclase (StlD) catalyzes the condensation of 5-phenyl-2,4-pentadienoyl-ACP and isovaleryl-beta-ketoacyl-ACP via a Michael addition. Such a pathway for stilbene formation is distinct from those widespread in plants. The Abe group solved the structure and biochemical mechanism of StlD and further investigated the aromatization reaction of the aromatase StlC. However, the generation of the required cinnamoyl-precursor 5-phenyl-2,4-pentadienoyl-ACP as a Michael acceptor for this cyclization reaction remained elusive. In this work, we were able to reconstitute the synthesis of the Michael acceptor in vitro, by the action of enzymes from the fatty acid biosynthesis. With the knowledge about the crucial cross-talk from primary and specialized metabolism, we further determined the minimal endowment for stilbene production in a heterologous host. Here, the discovered AasS enzyme StlB is responsible for the generation of cinnamoyl-ACP and among others, plFabH plays a key role as gatekeeper enzyme for further processing. With this information in hand, we were able to obtain IPS production in E. coli. These results are presented in the manuscript ‘Biosynthesis of the Multifunctional Isopropylstilbene in Photorhabdus laumondii Involves Cross-talk Between Specialized and Primary Metabolism’ (co-first author, manuscript).
The biosynthesis of the orange-to-red-pigmented anthraquinones (AQs) is the best-studied type II PKS system according to preliminary results. While several investigations by Brachmann et al. discovered the BGC and the overall product spectrum of the main AQ-256 and its methylated derivatives, data of Quiqin Zhou (Bode group) performed biochemical in vitro analysis paired with in vivo heterologous expression of the ant-genes antA-I. This led to the identification of shunt products that indicated an AQ-scaffold derived from an octaketide intermediate that gets shortened to a heptaketide by the hydrolase AntI, resulting in the main anthraquinone AQ-256. This PKS-shortening mechanism was further confirmed by the protein crystal structure of AntI by the Groll group (publication, minor contributions, co-author, Chem Sci. ‘Molecular Mechanism of Polyketide Shortening in Anthraquinone Biosynthesis of Photorhabdus luminescens’). Further substrate analysis of the P. luminescens AQ-producer and mutants revealed an inhibitory effect of cinnamic acid against the hydrolase AntI. Cinnamic acid might therefore be involved in regulation of AQ biosynthesis (‘Anthraquinone Production is Influenced by Cinnamic Acid’, first author, manuscript).
Biochemical analysis from Quiqin Zhou with the minimal PKS of the AQ-synthase further revealed the exclusive activation of the AQ-ACP by the PPTase AntB. The PPTase is insoluble alone but gets stabilized by the CoA-ligase, most likely inactive, working as a chaperone. Thus, the minimal PKS endowment to produce the octaketide scaffold compromises, besides the ACP, the KS:CLF heterodimer and the MCAT, the co-occurrence of the PPTase AntB and the CoA-ligase AntG. For the first time, X-ray crystallography depicted a minimal PKS in action, by obtaining the structural data of native complexes from an ACP:KS:CLF, the KS:CLF alone and an ACP:MCAT in their non-active and active forms. It was possible to confirm a KS-bound hexaketide, which was built upon heterologous expression of the KS:CLF. Mutagenesis with amino-acids proposed to be involved in protein-protein interactions in the ACP:KS:CLF complex revealed some interesting protein-interaction sites. Additionally, an induced-fit mechanism of the MCAT with the ACP during the malonylation reaction confirmed a monodirectional transfer reaction (‘Structural Snapshots of the Minimal PKS System Responsible for Octaketide Biosynthesis’ co-author, manuscript under review).
The metabolome of any live cell consists of several hundred, if not thousands of different molecules at any given moment, be it a relatively small bacterial cell or a whole multicellular organism. Although there are continuous attempts to differentiate between primary and secondary metabolites, the borders often blur in the eye of almost perfect interconvertability of all such matter. With chemistry and physics dominating this domain of biology it is an interdisciplinary endeavor to tackle the questions surrounding the workings of the metabolic pathways involved, searching for answers that ultimately help us to better understand life and find solutions to problems that affect us humans. One area of biochemistry that serves as a formidable example of the intertwined primary and secondary metabolic pathways are fatty acids, essential components of bacterial membranes, sources of energy and carbon but also important building blocks of several natural products. The second area to be mentioned is the metabolism of amino acids, the basic components of proteins and enzymes, which also serve as precursors to a diverse set of metabolites with many biological purposes.
This work focuses on these two areas of biochemistry, as several intermediates of their metabolism serve as building blocks for complex secondary metabolites whence many interesting and bioactive natural products are derived. The powerful and relatively novel tool of click-chemistry is employed to track azide-labeled precursors of primary and secondary metabolism in various bacterial strains to observe biochemistry at work and adds to the knowledge gained through other methods. The methods presented in this work serve the observation of fatty acid biosynthesis, degradation, modification and transport through direct ligation of azido fatty acids with cyclooctynes on one hand, leading to a revision of fatty acid transport in general. On the other hand a cleavable azide-reactive resin is devised to generally track the fate of azidated compounds through the myriads of metabolic pathways offered by entomopathogenic bacteria possessing a rich secondary metabolism. The resulting findings led to the identification of several antimicrobial peptides, amides and other compounds of which many had remained so far undetected in the strains that underwent investigation, underlining the worth of this method for future metabolomic research and beyond.
The metabolome of any live cell consists of several hundred, if not thousands of different molecules at any given moment, be it a relatively small bacterial cell or a whole multicellular organism. Although there are continuous attempts to differentiate between primary and secondary metabolites, the borders often blur in the eye of almost perfect interconvertability of all such matter. With chemistry and physics dominating this domain of biology it is an interdisciplinary endeavor to tackle the questions surrounding the workings of the metabolic pathways involved, searching for answers that ultimately help us to better understand life and find solutions to problems that affect us humans. One area of biochemistry that serves as a formidable example of the intertwined primary and secondary metabolic pathways are fatty acids, essential components of bacterial membranes, sources of energy and carbon but also important building blocks of several natural products. The second area to be mentioned is the metabolism of amino acids, the basic components of proteins and enzymes, which also serve as precursors to a diverse set of metabolites with many biological purposes.
This work focuses on these two areas of biochemistry, as several intermediates of their metabolism serve as building blocks for complex secondary metabolites whence many interesting and bioactive natural products are derived. The powerful and relatively novel tool of click-chemistry is employed to track azide-labeled precursors of primary and secondary metabolism in various bacterial strains to observe biochemistry at work and adds to the knowledge gained through other methods. The methods presented in this work serve the observation of fatty acid biosynthesis, degradation, modification and transport through direct ligation of azido fatty acids with cyclooctynes on one hand, leading to a revision of fatty acid transport in general. On the other hand a cleavable azide-reactive resin is devised to generally track the fate of azidated compounds through the myriads of metabolic pathways offered by entomopathogenic bacteria possessing a rich secondary metabolism. The resulting findings led to the identification of several antimicrobial peptides, amides and other compounds of which many had remained so far undetected in the strains that underwent investigation, underlining the worth of this method for future metabolomic research and beyond.
Myxobacteria are on order of Gram-negative, soil dwelling bacteria that feature an impressive number of properties: they can glide on solid surfaces by using two different motility motors, subsist by preying on other microorganisms, are often producers of multiple natural products, and upon adverse environmental conditions, they are able to form multicellular structures called “fruiting bodies”. The process, in which these macroscopically visible structures arise from independent single cells, has been the predominant subject of myxobacterial research for many decades. More precisely, researchers have strived for the discovery of genes, proteins and small molecules that act as signals, receivers or modulators of this complex process. In this regard, the species Myxococcus xanthus has evolved into the model organism due to its relatively simple and reliable handling in a laboratory environment. The research underlying this thesis focused on the identification and biosynthesis of lipids that may act as intercellular signaling molecules during the course of fruiting body formation of the myxobacterium Myxococcus xanthus as part of the “E-signal” system. In general, lipids containing branched-chain fatty acids with an uneven number of carbon atoms were found to be important players in this particular process. Nevertheless, their exact roles remain largely unknown as of this day. The first publication that is part of this thesis deals with an aspect that even strengthened the importance of role of iso-branched compounds in myxobacteria: myxobacterial metabolism is able to transform precursors of iso-lipids to isoprenoids. It addresses the question whether isoprenoids in general are important for fruiting body formation. Phenotypic analysis of mutants impaired in the biosynthesis of the central isoprenoid precursor 3-hydroxymethylglutaryl-Coenzyme A (3-HMG-CoA) from acetate and/or branched chain keto acids and their genetic and metabolic complementation clearly showed that isoprenoids are essential for fruiting body formation and confirmed that leucine derived isovalerate is an important source for isoprenoid precursors in myxobacteria. The second, and by far and away most tedious and sophisticated study, addressed the question as to how myxobacteria form fatty acid derived iso-branched ether lipids and to what extent they are important for fruiting body formation and sporulation. In a previous study, those unusual lipids were identified as specific biomarkers for myxobacterial development. No biochemical pathways to ether lipids specific for prokaryotes were known by then. In this study, a putative candidate gene that may be in involved in ether lipid biosynthesis was investigated. A combination of gene disruption and complementation experiments, phenotypic analysis and monitoring of ether lipid formation by means of GC-MS demonstrated its involvement in myxobacterial ether lipid biosynthesis and the importance of these lipids for the developmental process. Heterologous expression and biochemical testing of this gene together with in-silico sequence analysis and docking experiments confirmed the functions of its predicted domains. The discussion section provides an additional suggestion on how the ether bond formation is performed. Furthermore and most importantly, iso-branched ether lipids were found to be essential for sporulation but not for fruiting body formation. In summary, one or several molecules derived from an iso-branched alkylglycerol seem to play a role during sporulation in M. xanthus and a multidomain enzyme unique for myxobacteria is involved in their biosynthesis. The last manuscript addresses the complexity of lipid metabolism in myxobacteria. Prior to this work, there was limited knowledge about the exact composition of the myxobacterial lipidome and no method was available to monitor putative changes in the myxobacterial lipidome down to the single molecular species for studying lipid biosynthesis or regulation. An ultra-performance liquid chromatography coupled with mass spectrometry based method with electrospray ionization (UPLC-ESI-MS) utilizing standard equipment and a water/acetonitrile/isopropanol based eluent system proved to be geared for the construction of lipid profiles for wild type and mutant cells of M. xanthus and to show their differences. Fragmentation spectra based structure elucidation of lipid molecular species resulted in the identification of 99 molecular species comprising glycerophosphoethanolamines, glycerophosphoglycerols, glycerolipids, ceramides and ceramide phosphoinositols. The latter have never been described for any prokaryotes before. Three dimensional plots were created from the relative intensity differences of the single molecular ion species between the different samples to provide an efficient and versatile visualization of the data and enable the researcher to quickly detect differences.
Nematophilic bacteria as a source of novel macrocyclised antimicrobial non-ribosomal peptides
(2020)
A solution to ineffective clinical antimicrobials is the discovery of new ones from under-explored sources such as macrocyclic non-ribosomal peptides (NRP) from nematophilic bacteria. In this dissertation an antimicrobial discovery process –from soil sample to inhibitory peptide– is demonstrated through investigations on six nematophilic bacteria: Xenorhabdus griffiniae XN45, X. griffiniae VH1, Xenorhabdus sp. nov. BG5, Xenorhabdus sp. nov. BMMCB, X. ishibashii and Photorhabdus temperata. To demonstrate the first step of bacterium isolation and species delineation, endosymbionts were isolated from Steinernema sp. strains BG5 and VH1 that were isolated directly from soil samples in Western Kenya. After genome sequencing and assembly of novel Xenorhabdus isolates VH1 and BG5, species delineation was done via three overall genome relatedness indices. VH1 was identified as X. griffiniae VH1, BG5 as Xenorhabdus sp. nov. BG5 and X. griffiniae BMMCB was emended to Xenorhabdus sp. nov. BMMCB. The nematode host of X. griffiniae XN45, Steinernema sp. scarpo was highlighted as a putative novel species. To demonstrate the second step of genome mining and macrocyclic non-ribosomal peptide structure elucidation, chemosynthesis and biosynthesis, the non-ribosomal peptide whose production is encoded by the ishA-B genes in X. ishibashii was investigated. Through a combination of refactoring the ishA-B operon by a promoter exchange mechanism, isotope labelling experiments, high resolution tandem mass spectrometry analysis, bioinformatic protein domain analysis and chemoinformatic comparisons of actual to hypothetical mass spectrometry spectra, the structures of Ishipeptides were elucidated and confirmed by chemical synthesis. Ishipeptide A was a branch cyclic depsidodecapeptide macrocyclised via an ester bond between serine and the terminal glutamate. It chemosynthesis route was via a late stage macrolactamation and linearised Ishipeptide B was synthesised via solid phase iterative synthesis. Ishipeptides were not N-terminally acylated despite being biosynthesised from the IshA protein that had a C-starter domain. It was highlighted that more than restoration of the histidine active site of this domain is required to restore N-terminal acylation activity.
To demonstrate the final step of determination of antimicrobial activity, minimum inhibitory concentrations of Ishipeptides and Photoditritide from Photorhabdus temperata against fungi and bacteria were determined. None were antifungal while only the macrocyclic compounds were inhibitory, with Ishipeptide A inhibitory to Gram-positive bacteria at 37 µM. The cationic Photoditritide, a cyclic hexapeptide macrocyclised via a lactam bond between homoarginine and tryptophan, was 12 times more inhibitory (3.0 µM), even more effective than a current clinical compound, Ampicillin (4.2 µM). For both, macrocyclisation was hypothesised to contribute to antimicrobial activity. Ultimately, this dissertation demonstrated not only nematophilic bacteria as a source of novel macrocyclic antimicrobial non-ribosomal peptides but also a process of antimicrobial discovery–from soil sample to inhibitory peptide– from these useful bacteria genera. This is significant for the fight against antimicrobial resistance.
This work comprises the investigation of four different biosynthesis gene clusters from Xenorhabdus. Xenorhabdus is an entomopathogenic bacterium that lives in mutualistic symbiosis with its Steinernema nematode host and together they infect and kill insect larvae. Xenorhabdus is well known for the production of so-called specialised metabolites and many of these compounds are synthesised by non-ribosomal peptide synthetases (NRPSs) or NRPS-polyketide synthase (PKS)-hybrids. These enzymes are organised in a modular manner and produce structurally very diverse molecules, often with the help of modifying domains and tailoring enzymes. In general, the genes involved in the biosynthesis are organised in so-called biosynthetic gene clusters (BGCs) in the genome of the producing strain. Exchanging the native promoter with an inducible promoter, e.g. PBAD, allows the targeted activation of the BGC and in turn the analysis of the biosynthesis product via LC-MS analysis.
The first BGC investigated in this work is responsible for the biosynthesis of xenofuranones. Based on gene deletions, this work shows that the NRPS-like enzyme XfsA produces a carboxylated furanone intermediate which is subsequently decarboxylated by XfsB to yield xenofuranone B. The next step in xenofuranone biosynthesis is the O-methylation of xenofuranone B to yield xenofuranone A. A comparative proteomics approach allowed the identification of four methyltransferase candidates and subsequent gene deletions confirmed one of the candidates to be responsible for methylation of xenofuranone B. The proteome analysis was based on the comparison of X. szentirmaii WT and X. szentirmaii Δhfq because distinct levels of the methylated xenofuranone A were observed when the xfs BGC was activated in either WT or Δhfq strain. Hfq is a global transcriptional regulator whose deletion is associated with the down regulation of natural product biosynthesis in Xenorhabdus. The strong PBAD activation of the xfs BGC also allowed the detection of two novel xenofuranone derivatives which arise from incorporation of one 4-hydroxyphenylpyruvic acid as first or second building block, respectively.
PBAD based activation of the second BGC addressed in this work lead to the detection of a novel metabolite and compound purification allowed NMR-based structure elucidation. The molecule exhibits two pyrrolizidine moieties and was named pyrrolizwilline (pyrrolizidine + twin (German: “Zwilling”)). The BGC comprises seven genes and single gene deletions as well as heterologous expression in E. coli and NRPS engineering were conducted to investigate the biosynthesis. The first two genes xhpA and xhpB encode a bimodular NRPS and a monooxygenase which synthesise a pyrrolizixenamide-like structure, similar to PxaA and PxaB in pyrrolizixenamide biosynthesis. It is suggested that the acyl side chain incorporated by XhpA is removed by the α,β-hydrolase XhpG. The keto function is then reduced by two subsequent two electron reductions catalysed by XhpC and XhpD. One of these two reduced pyrrolizidine units most likely is extended with glyoxalate prior to non-enzymatic dimerisation with the second pyrrolizidine moiety. To finally yield pyrrolizwilline, L-valine is incorporated, probably by the free-standing condensation domain XhpF.
The third BGC investigated is responsible for the production of a tripeptide composed of β-D-homoserine, α-hydroxyglycine and L-valine and is referred to as glyoxpeptide. This work demonstrates that the previously observed glyoxpeptide derivative is derived from glycerol present in the culture medium. Furthermore, this work shows that the monooxygenase domain, which is found in an unusual position between motifs A8 and A9 within the adenylation domain, is responsible for the α-hydroxylation of glycine. It is suggested that the α-hydroxylation of glycine renders the tripeptide prone to hydrolysis via hemiacetal formation. Hence, the XgsC_MonoOx domain might be an interesting candidate for further NRPS engineering.
The fourth BGC addressed is responsible for the production of xildivalines and this work describes two additional derivatives which are detected only when the promoter is exchanged and activated in the X. hominickii WT strain but not in X. hominickii Δhfq. Deletion of the methyltransferase encoding gene xisE results in the production of non-methylated xildivalines. It remains to be determined when the N-methylation of L-valine takes place. It is discussed that the methyltransferase could act on the NRPS released product but also during the assembly. The peptide deformylase is not involved in the proposed biosynthesis as xildivaline production is detected in a ΔxisD strain. The PKS XisB features two adjacent, so-called tandem T domains. The inactivation of the first or the second T domain by point mutation causes decreased production titres of detected xildivalines in the respective mutant strain when compared to the wild type.
Non-ribosomal peptide synthetases (NRPSs) are modular biosynthetic megaenzymes producing many important natural products and refer to a specific set of peptides in bacteria’s and fungi’s secondary metabolism. With the actual purpose of providing advantages within their respective ecological niche, the bioactivity of the structurally highly diverse products ranges from, e.g., antibiotic (e.g., vancomycin) to immunosuppressive (e.g., cyclosporin A) to cytostatic (e.g., echinomycin or thiocoralin) activity.
An NRPS module consists of at least three core domains that are essential for the incorporation of specific substrates with the 'multiple carrier thiotemplate mechanism' into a growing peptide chain: an adenylation (A) domain selects and activates a cognate amino acid; a thiolation (T) domain shuffles the activated amino acid and the growing peptide chain, which are attached at its post-translationally 4ʹ-phosphopantetheine (4'-PPant) group, between the active sites; a condensation (C) domain links the upstream and downstream substrates. NRPS synthesis is finished with the transfer of the assembled peptide to the C-terminal chain-terminating domain. Accordingly, the intermediate is either released by hydrolysis as a linear peptide chain or by an intramolecular nucleophilic attack as a cyclic peptide.
The NRPS’s modular character seems to imply straightforward engineering to take advantage of their features but appears to be more challenging. Since the pioneering NRPS engineering approaches focused on the reprogramming and replacement of A domains, several working groups developed advanced methods to perform a complete replacement of subdomains or single or multiple catalytic domains.
The first part of this work focusses parts of the publication with the title 'De novo design and engineering of non-ribosomal peptide synthetases', which follows up assembly line engineering with the development of a new guideline. Thereby, the pseudodimeric V-shaped structure of the C domain is exploited to separate the N-terminal (CDSub) and C-terminal (CASub) subdomains alongside a four-AA-long linker. This results in the creation of self-contained, catalytically active CASub-A-T-CDSub (XUC) building blocks. As an advantage over the previous XU concept, the characteristics (substrate- and stereoselectivity) assigned to the C domain subunits are likewise exchanged, and thus, no longer represent a barrier. Furthermore, with the XUC concept, no important interdomain interfaces are disrupted during the catalytic cycle of NRPS, allow to expect much higher production titers. Moreover, the XUC concept shows a more flexible application within its genus origin of building blocks to create peptide libraries. Additionally, with this concept only 80 different XUC building blocks are needed to cover the entire proteinogenic amino acid spectrum.
The second part of this work addresses the influence of the C domain on activity and specificity of A domains. In a comprehensive analysis, a clear influence of different C domains on the in vitro activation rate and the in vivo substrate spectrum could be observed. Further in situ and in silico characterizations indicate that these influences are neither the result of the respective A domains promiscuity nor the C domain’s proofreading, but due to an 'extended gatekeeping' function of the C domain. This novel term of an 'extended gatekeeping' function describes the very nature of interfaces that C domains can form with an A domain of interest. Therefore, the C-A interface is assumed to have a more significant contribution to a selectivity filter function.
The third part of this work combines the NRPS engineering with phylogenetic/evolutionary perspectives. At first, the C-A interface could be precisely defined and further identified to encode equivalent information corresponding to the complete C-A didomain. Moreover, the comparison of NRPSs topology reveals hints for a co-evolutionary relatedness of the C-A didomain and could be shown to reassemble even after separation. In this regard, based on a designed CAopt.py algorithm, the reassembling-compatibility of hybrid interfaces could be determined by scoring of the co-expressed NRPS hybrids. This algorithm also enables the randomization of the interface sequences, thus, leading to the identification of more functional interface variant, which cause significantly higher peptide production and could even be applied to other native and hybrid interfaces.
Die Physiologie des Schmerzes umfasst komplexe immunologische, sensorische und inflammatorische Prozesse im Rückenmark, im Gehirn und in der Peripherie. Wiederholte nozizeptive Stimulation induziert pathophysiologische Veränderungen bei der Schmerzweiterleitung, aus denen eine periphere oder zentrale Sensibilisierung resultiert. Diese kann bei dafür anfälligen Patienten zu der Ausbildung von chronischen Schmerzzuständen führen. Obwohl das Wissen über die genauen molekularen Vorgänge der Schmerz-Chronifizierung noch immer unvollständig ist, sind die Identifizierung von Risikofaktoren vernünftige Schritte, um die individuelle Anfälligkeit für die Entwicklung chronischer Schmerzen zu bestimmen. Das Hauptziel dieser Doktorarbeit bestand daher in der Identifikation humaner genetischer Biomarker für chronische Schmerzzustände.
In the 'Golden Age of Antibiotics', between 1940 and 1970, the global pharmaceutical companies discovered many antibiotics, such as cephalosporins, tetracyclines, aminoglycosides, glycopeptides, etc., as well as antifungal and antiparisitic agents. Due to several reasons, e.g. the steady re-discovery of already known NPs and the associated high costs, many pharmaceutical companies have significantly scaled back or totally abandoned their NP discovery programs since the late 20th century. Instead those companies started to focus on drug discovery based on combinatorial synthesis and thereby on the creation of enormous synthetic libraries containing small molecules. Unfortunately, this synthetic approach dealing with the optimization of existing NP or antibiotic has its limitations. As a result, leading pharmaceutical companies are re-conducting NPs research to discover new antimicrobials for the upcoming antimicrobial resistance threat. The Natural Product Center of Excellence, a collaboration between Sanofi-Aventis and Fraunhofer IME, is advancing in this context the discovery and development of novel antimicrobial agents for the treatment of infectious diseases through the testing of Sanofi's microbial extract library and strain collection. The aim of the present PhD thesis was the discovery and isolation of novel antimicrobial compounds with improved activities and/or novel MOAs as potential lead compound for a further drug discovery.
The compound class of the fabclavines was described as secondary or specialized metabolites (SM) for Xenorhabdus budapestensis and X. szentirmaii. Their corresponding structure was elucidated by NMR and further derivatives could be identified in both strains. Biochemically, fabclavines are hybrid SMs derived from two non-ribosomal-peptide-synthetases (NRPS), one type I polyketide-synthase (PKS) and polyunsaturated fatty acid (PUFA) synthases. In detail, a hexapeptide is connected via partially reduced polyketide units to an unsual polyamine. Structurally, they are related to the (pre-)zeamines, described for Serratia plymuthica and Dickeya zeae. Fabclavines exhibit a broad-spectrum bioactivity against a variety of different organisms like Grampositive and Gram-negative bacteria, fungi, protozoa but also against eukaryotic celllines.
In this work, the fabclavine biosynthesis was elucidated and assigned to two independently working assembly lines. The NRPS-PKS-pathway is initiated by the first NRPS FclI via generation of a tetrapeptide, which is elongated by the second NRPS FclJ, leading to a hexapeptide. Alternatively, FclJ can also act as direct start of the biosynthesis, resulting in the final formation of shortened fabclavine derivatives with a diinstead of a hexapeptide. In both cases, the peptide moiety is transferred to the iterative type I PKS FclK, leading to an elongation with partially reduced polyketide units. The resulting NRPS-PKS-intermediate is still enzyme-bound. The PUFA-homologues FclC, FclD and FclE in combination with FclF, FclG and FclH belong to the polyamine-forming pathway. Briefly, repeating decarboxylative Claisen thioester condensation reactions of acyl-coenzym A building blocks lead to the generation of an acyl chain in a PKS- or fatty acid biosynthesis-like manner. The corresponding β-keto-groups are either completely reduced or transaminated in a specific and repetitive way, resulting in the concatenation of so-called amine-units. The final β-keto-group is reduced to a hydroxy-group and the intermediate is reductively released by the thioester reductase FclG. A subsequent transamination step leads to the final polyamine. The NRPS-PKS- as well as the polyamine-pathway are connected by FclL. This condensation domain-like protein catalyzes the condensation of the polyamine with the NRPS-PKS-part, which results in the release of the final fabclavine. The results are described in detail in the first publication (first author).
Fabclavine biosynthesis gene cluster (BGC) are widely spread among the genus Xenorhabdus and Photorhabdus. In Xenorhabdus strains a high degree of conservation regarding the BGC synteny as well as the identity of single proteins can be observed. However, Photorhabdus strains harbor only the PUFA-homologues. While in Photorhabdus no product could be detected, our analysis revealed that the Xenorhabdus strains produce a large chemical diversity of different derivatives. Briefly, the general backbone of the fabclavines is conserved and only four chemical moieties are variable: The second and last amino acids of the NRPS-part, the number of incorporated polyketide units as well as the number of amine units in the polyamine. In combination with the elucidated biosynthesis, these variables could be assigned to single biosynthesis components as diversity mechanisms. Together with the 10 already described derivatives, a total of 32 derivatives could be detected. Interestingly, except for taxonomic closely related strains, all analyzed strains produce their own set of derivatives. Finally, we could confirm that the fabclavines are the major bioactive compound class in the analyzed strains under laboratory conditions. The results are described in detail in the second publication (first author).
Together with our collaboration partner Prof. Selcuk Hazir a potent bioactivity against Enterococcus faecalis, which is associated with endodontic infections, could be contributed to X. cabanillasii. Here, we could confirm that this bioactivity can be assigned to the fabclavines. The results are described in detail in the third publication(co-author).
Among the genus Xenorhabdus, X. bovienii represents an exception as its NRPS and PKS genes of the fabclavine BGC are missing or truncated, resulting in the exclusive production of polyamines. Furthermore, its PUFA-homologue FclC harbors an additional dehydratase (DH) domain. Upon extensive analysis a yet unknown deoxy-polyamine was identified and assigned to this additional domain. Finally, the DH domain was transferred into other polyamine pathways. Regardless of an in cis or in trans integration, the chimeric pathways produced deoxy-derivatives of its naturally occurring polyamines, suggesting that this represents another diversification mechanism. The results are described in detail in the attached manuscript (first author).
This work characterizes the post-PKS modifications of AQ-256. Additionally, the second part describes the establishment of an AQ production platform for electrolyte generation that can be utilized in redox-flow-batteries. Lastly, a silent BGC that encodes the genes for terpenoid biosynthesis was described and characterized with regards to product formation and putative ecological function.
Identification of new natural products from nematode-associated bacteria using mass spectrometry
(2023)
This work aims to find unknown natural products produced by bacteria, that live in close association with nematodes and to elucidate their structure by using mass spectrometry.
The first chapter of this work is dedicated to the detection of hitherto unknown natural products by using a metabolomics approach and subsequent structure elucidation of said compounds. This chapter includes metabolomics analysis of Xenorhabdus szentirmaii wild type and knockout mutants, overproduction of the target compound, identification of derivatives from other strains and MS based structure elucidation.
The second and third chapters are about natural products that protect C. elegans from B. thuringiensis infections.
The second chapter deals with natural products that protect the nematode host without killing the pathogen. I deployed molecular biology methods to generate deletion and overproduction strains of a target compound, identified it via LC-MS/MS analysis and used LC-MS/MS and lipidomics to analyse the chemical properties of the active compound.
The third chapter aims at finding natural products, which are produced by Pseudomonas strains MYb11 and MYb12, respectively. These natural products display the ability to protect C. elegans by killing B. thuringiensis. I identified said compounds via fractionation and subsequent bioactivity testing. After identification, I generated production strains of the target compounds and elucidated the structure of the bioactive derivative.
The last chapter deals with the structure elucidation of peptides produced by an unusual GameXPeptide synthetase in Xenorhabdus miraniensis. I analysed producer strains of GameXPeptides using LC-MS and elucidated the structural differences between the known GameXPeptides, produced by P. luminescens TT01, and the unusual ones produced by X. miraniensis.
This work addresses the investigation of the biosynthesis mechanisms of type II polyketide synthase (PKS) and fatty acid synthase (FAS) derived specialized metabolites (SMs) from Photorhabdus laumondii.
The elucidation of the biosynthetic pathway of the bacterial 3,5-dihydroxy-4-isopropyl-trans-stilbene (IPS) was one of the major topics of this thesis. IPS exhibits several bioactive characteristics as it inhibits the phenoloxidase of insects, acts antibacterial, but also influences the soluble epoxide hydrolase which is involved in inflammatory reactions. It was recently approved as a treatment against psoriasis by the FDA and is the first Photorhabdus derived drug.
The stilbene generation in Photorhabdus requires the formation of the two acyl-carrier-protein (ACP) bound 5-phenyl-2,4-pentadienoyl- and isovaleryl-β-ketoacyl-moieties. The ketosynthase (KS)/cyclase StlD catalyzes a ring formation via a Michael-addition between the two intermediates which is then further processed by an aromatase. The formation of 5-phenyl-2,4-pentadienoyl-ACP was shown via in vitro assays with purified proteins by proving the influence of the KS FabH, ketoreductase FabG and dehydratase FabA or FabZ of the fatty acid metabolism. While E. coli was able to complement most of these enzymes in attempts to produce IPS in the heterologous host, the Photorhabdus derived FabH was not replaceable despite 73 % sequence identity with the E. coli based isoenzyme, acting as a gatekeeper enzyme for cinnamic acid (CA) moieties. Furthermore, the ability to incorporate meta-substituted halogenated CA-derivatives was shown in order to produce 3-chloro- and 3-bromo-IPS. While studying the stilbene biosynthesis, the ability of Photorhabdus and Xenorhabdus to produce hydrazines was also discovered.
The second investigated biosynthesis was the formation of benzylideneacetone (BZA). BZA is produced by Photorhabdus and Xenorhabdus strains acting as a suppressor for the immune cascade of insects and has also antibiotic activities towards Gram-negative bacteria. Due to its structural similarity towards CA and the intermediates during the stilbene formation, a shared mechanism for Photorhabdus and Xenorhabdus budapestensis was proposed due to their ability to produce CA. The production of BZA was also dependent on the stilbene related CoA-ligase, the ACP and FabH. It was verified in vitro and in vivo in E. coli yielding a 150-fold increase of the BZA production compared to the Photorhabdus and Xenorhabdus wildtype (WT) strains.
The second part of this work deals with the optimization of P. laumondii strains regarding the production titers of IPS. Therefore, several deletions of other SM related genes as well as promoter exchanges in front of stilbene related genes were carried out. These approaches were combined with the upregulation of the phenylalanine by heterologous plasmid expression, since it is the precursor of CA. Another approach applied in parallel was the optimization of the cultivation conditions with different media and supplementation with XAD-resins. It was proved that media rich on fatty acids or peptides led to higher optical densities of the cultures and thus to higher titers of stilbenes. Since IPS is inhibiting the phenoloxidase, an enzyme important for the insect immunity, it was hypothesized that cultivation in media containing insects might enhance the output of this SM. Starting from 23 mg/l of IPS in the P. laumondii WT strain, it was possible to increase the production levels to more than 860 mg/l by utilizing the mentioned approaches.
The last topic of this thesis focuses on the production of epoxidated IPS (EPS) and its derivatives. Under laboratory conditions, only a low titer of EPS was observed for the wildtype strain. However, the optimized IPS strains and IPS-production conditions could also be applied for EPS which led to higher productions and also to the detection of many new derivatives. Most of the EPS derivatives were amino acid or peptide derived acting as nucleophiles to open the epoxide ring and yielding β-amino-alcohols. However, purification and chemical synthesis attempts to obtain EPS failed due to its poor stability. Epoxides were utilized in in vitro assays with amino acids, peptides and proteins to get insights whether epoxidations might act as posttranslational modification in Photorhabdus. The reactions were performed with styrene oxide and stilbene oxide replacing EPS based on their structural similarity. The modifications were executed successfully although proteomics approaches with in vivo data are required to confirm these findings. During the purification attempts of EPS, further derivatives were detected. The structures of dimerized stilbenes, a cis-isomer of IPS and another derivative that might incorporate an amino-group in the resveratrol ring were proposed on the basis of the HPLC-MS data.