Year: 2025
Type: dissertation
Abstract: Natural products (NP) are characterised by a high degree of chemical diversity and complexity. These substances offer a variety of functions for the produced organisms, which are typically bacteria, fungi and plants. These compounds frequently serve as a source for the development of therapeutic agents, which in turn have numerous applications, including antibiotics and anticancer agents. A considerable proportion of these NPs are synthesised by non-ribosomal peptide synthetases (NPRS), which are large megaenzymes defined by their characteristic modular architecture. These modules can be categorised into three distinct types: starter/initiation, elongation and termination modules. The fundamental module architecture comprises an adenylation (A) domain, which is responsible for the selection and activation of an amino acid, a thiolation (T) domain, which covalently binds the activated amino acid and mediates the interaction of the substrate with other domains, and a condensation (C) domain, which is responsible for the condensation of the activated amino acid with the upstream peptide chain. The thioesterase (TE) domain, which is found in termination modules, is responsible for the release of the peptide from the enzyme. Further alterations and modifications may encompass epimerisation, hererocyclisation, methylation, oxidation, acylation, and reduction. These additional processes are usually executed by further domains, thereby enhancing the peptide’s structural complexity. The modular architecture of NRPS has inspired the development of numerous NRPS engineering strategies over the past few decades. This encompasses mutagenesis and the exchange of domains and subdomains. In addition to them, module exchange strategies such as the “eXchange Unit” (XU) concept, the “eXchange Unit Condensation domain” (XUC) concept, and the recently developed evolution-inspired “eXchange Unit between T domains” (XUT) concept have proven to be quite successful. These strategies are employed in the active engineering of NRPS systems, with the objective of modifying and altering the resulting NP. Hence, this process provides biotechnological opportunities for the creation of NPs and derivatives of interest. The assembly of engineered NRPS systems is typically accomplished through fragment-based assembly strategies, such as Gibson assembly. However, despite the efficiency of fragment-based assembly strategies in conjunction with modular exchange strategies, these strategies frequently exhibit deficiencies for effective high-throughput applications. Among a multitude of diverse assembly strategies, the Golden Gate Assembly (GGA) has been employed for a wide range of applications in the field of genetic fragment assembly and library creation. Depending on the genetic environment and the organisms targeted, different limitations must be considered. Consequently, a wide range of “syntaxes” for GGA have been developed. These syntaxes typically encompass preferred type IIS restriction enzymes, functional overhangs and acceptor and donor systems. The main project of this thesis was the development of a GGA syntax for NRPS engineering. The GGA strategy provides a range of advantages, including a scarless assembly method that can be leveraged using the XUT approach for NRPS engineering. It is evident that among the multiple fusion sites offered by this engineering strategy, the XUTI and XUTIV fusion sites yielded the most favourable results. The XUTIV fusion site, located within the conserved motif FFxxGGxS, can be utilised for the creation of overhangs without altering the amino acid sequence, thereby providing a scarless application for GGA. The conserved glycines were employed to generate 16 overhangs, based on their codons. The viability of the 16 overhangs was assessed in silico using fidelity tools, which reduced them to smaller sets containing six to nine highly compatible overhangs. Hence, using the designated overhangs, donors containing XUTs, as well as acceptors of interest, were designed. As a proof of concept, individual controls were assembled using the GGA as well as the alternative Gibson assembly. Subsequently, a library of donors was generated and applied to create a tri- and tetra-modular NRPS library, containing various elongation modules in the form of XUTs. Furthermore, an efficient qPCR-based method was developed for the efficient confirmation of NRPS constructs within a GGA-based NRPS library. In addition to elongation modules, the systems were adapted to include starter modules, which resulted in the creation of an additional starter module library. Moreover, the developed systems were employed not only for the generation of novel NRPS systems, but also for the purpose of active derivatisation. Hence, the native xenoamicin synthetase (XabABC) was adapted as an acceptor to target the cyclic structure of the resulting depsipeptides by modifying its amino acid sequence and size. Consequently, an efficient syntax for GGA-based NRPS engineering has been established and applied for the generation of over 100 novel NRPS systems, allowing for the modification of starter, elongation, and termination modules, respectively. Additionally, the active derivatisation of the XabABC NRPS resulted in the generation of over 50 different systems and 25 novel xenoamicin derivatives (“High-Throughput Engineering and Modification of Non-ribosomal Peptide Synthetases Based on Golden Gate Assembly”, submitted to Angewandte Chemie & “Peptide Libraries, Methods and Protocols”, shared-first author, Methods Mol. Biol. 2025). As a second topic, the NRPS-A1_D3 was investigated, which was created during the GGA-based NRPS engineering development and showed macroscopic observable aggregation and sedimentation of the producing strain during production. This aggregation was studied using growth assays as well as microscopical imaging. For quantification, a crystal violet assay was performed. However, despite significant sedimentation and aggregation, the quantification was not successful. Concurrently, a position scan for the expected peptide and derivatives was conducted, thereby confirming the necessity of the N-terminal fatty acid and at least one of the provided aromatic amino acids for the induction of aggregation. As a third topic, the developed GGA-based NRPS engineering strategy was employed for the structure elucidation of unique building blocks of interest. The unique database provided by NPRS libraries offers valuable information for the retention times of similar compounds, but especially for the comparison of fragmentation patterns. Therefore, it is possible to elucidate unique building blocks based on the exclusion of fragmentations, which are consistently and reliably observed within a library. This was firstly applied for a novel non-ribosomal peptide synthetase/polyketide synthase (NRPS/PKS) engineering strategy, which utilised PKS elongation modules comprising peptidyl carrier protein (PCP) domains at both termini. Consequently, these distinctive PKS modules can be engineered through the utilisation of the XUT concept and employed in conjunction with GGA. Hence, the aetokthonostatin (AEST)- producing biosynthetic gene cluster (BGC) aesABCDEFGHIJK from Aetokthonos hydrillicola was employed, providing two XUT-compatible PKS modules applicable for this strategy. From the resulting hybrid constructs, two anticipated compounds were detected and the comparison of fragmentations, as well as feeding experiments, demonstrated that the remaining mass correlated with PKS building blocks, such as malonyl- and methylmalonyl moieties. To achieve definitive elucidation, a collaboration was initiated with the research group led by Professor Dr. Timo Niedermeyer from the University of Berlin. Secondly, this elucidation strategy was employed to investigate a unique module from a putative BGC (dindS) found in Shewanella violacea, which produces an indigoidinelike compound. This module contains a reductase (R) domain instead of a typical C domain, thereby catalysing the reduction of the carboxy group of the upstream moiety. Consequently, in contrast to the typical peptide bond, an amine group is observed. This module was therefore used as an elongation module, along with several acceptors, to examine whether the alteration could take place between two amino acids rather than a fatty acid and an amino acid, as observed in the native system. However, despite the observation of a compound with the anticipated peptide sequence, the expected amine-containing modification was not observed. The final subject of this thesis was the investigation of cysteine-incorporating NRPS modules. Thus, a recombinant NRPS system, which was previously created in the author’s master thesis, was observed to produce two distinct compounds that resembled decapeptides being formed via a disulfide bond from cysteines. However, the particular structure of these compounds and thus the incorporation of cysteines was not elucidated in the master thesis. Consequently, in this work, a replication of the reported feeding experiments was conducted, confirming the results. To elucidate the structure of the two compounds, they were purified and utilised in conjunction with a reducing agent. This confirmed the presence of cysteines by providing two pentapeptides with the expected peptide sequence after reduction. The confirmed cysteine-incorporating module was implemented into the GGA-based NRPS engineering environments in the form of donors. However, following the assembly of various systems, no expected cysteine-containing compounds could be observed. Finally, an alternative approach was adopted, involving the deactivation of the cyclisation reaction through the introduction of specific mutations within the responsible catalytic centre. However, while the elimination of production for the heterocyclised compound was observed as expected, the anticipated derivative, which should contai
Author: Podolski, Adrian
Institution: Humboldt-Universität zu Berlin
Language: English
Subfield: Applied Microbiology and Biotechnology
Field: Immunology and Microbiology
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7225366033
