Abstract:
Most of todays applied antibiotics belong to the group of natural products or have been derived from them, by modifying individual functional groups. In the search for new antibiotics, current research is again concentrating on natural product discovery, since those compounds combine complex physiochemical properties with regard to good bacterial uptake, intracellular retainment, and target engagement, which cannot simply be mimicked by rational design. There is justified hope that the investigation of new producer strains from less explored niches could reveal new natural products. Genome sequencing has also shown that many producer strains possess the potential to synthesize additional secondary metabolites, which are encoded in so-called cryptic or silent gene clusters. The activation of those gene clusters, e.g., by cultivation of the producer strains under different growth or stress conditions, genetic manipulation of gene regulation, or heterologous expression, is also part of current research. In order to effectively screen all those potential natural product producers, improved purification and dereplication methods are required and, above all, fast, robust, and selective screening procedures, which generate high content information.
In this thesis, an agar-based bioreporter approach was developed and validated, which allows for combined bioactivity and mode of action-informed screening. The employed
ß-galactosidase-based Bacillus subtilis bioreporter constructs showed selective induction upon antimicrobial interference with one of the main metabolic pathways: DNA synthesis (PyorB-lacZ), RNA synthesis (Prpt-lacZ and PhelD-lacZ), protein biosynthesis (PbmrC-lacZ, selective for translation arrest) and integrity of the cell envelope (PypuA-lacZ and PliaI-lacZ). Induction specificity of the bioreporter panel in the agar-based setup was confirmed using
a large set of ~90 reference antibiotics with known MOA. Of note, Prpt had not previously been described as bioreporter and was therefore extensively profiled for its induction specificity and sensitivity upon RNA stress in the agar-based assay as well as in a liquid, luciferase-based system. After validation of the bioreporter panel, the agar-based approach allowed the characterization of unknown antimicrobial agents, like the investigated compounds microcionamide A and C. Generally, the bioreporter-based MOA profiling indicated the metabolic pathway(s) of antibiotic interference and allowed for the selection of adequate MOA follow-up studies to elucidate the exact molecular target.
Furthermore, the agar-based bioreporter approach was evaluated for its applicability in direct screening of antibiotic producer strains. It could be proven that produced antimicrobials can be sensitively detected and MOA-profiled by direct probing of agar plugs or culture supernatants of the cultivated producer strains, with no need for initial compound purification. This finding was especially interesting as antibiotic producer strains often possess the genetic potential to synthesize different antimicrobial secondary metabolites, which can rapidly be distinguished using the additional MOA information of the bioreporter panel. The assay also bears the advantage to monitor the entire purification process of
a bioactive substance, allowing to test unprocessed producer strains, extracts, fractionated samples, or pure substances for the presence of the previously detected activity and bioreporter signal of interest. It could further be shown, that the agar-based bioreporter approach is able to sensitively detect polypharmacology or synergistic effects of antimicrobial substances.
The specific upregulation of the selected B. subtilis bioreporter genes upon antibiotic interference with one of the upper mentioned main metabolic pathways was discovered in a previous transcriptome study. However, the function and regulation of the identified
B. subtilis genes are only partially characterized. A better understanding of the initiated stress responses that regulate selective gene induction would enable a refined
MOA characterization of the inducing antibiotics, and potentially allow new insights into cellular adaptation or resistance mechanisms. Hence, regulation of the bioreporter genes rpt, helD, and yorB, which are selectively upregulated upon exposure to antibiotics that inhibit RNA or DNA synthesis, respectively, was investigated in this work. For gene regulation of rpt and helD, the induction profiles of Prpt and PhelD upon antibiotic treatment in a quantitative, luciferase-based reporter system, already pointed at a similar regulation mechanism. Further results like promoter deletion studies or DNA affinity capturing assays indicated, that the SigA-regulated genes rpt and helD both belong to a cis-encoded antisense RNA-regulated stress response, which is proposed to be selectively induced by the cellular sensing of stalled RNA polymerase complexes. The SPß-prophage protein YorB was proven to be LexA-regulated and thus part of the SOS response in B. subtilis. Finally, the function of B. subtilis Rpt (formerly PpS/YppS), which had not been previously characterized, was elucidated. It could be shown that Rpt confers selective resistance to rifamycins by phosphorylation inactivation.
