Biochemistry
Identification and characterization of small molecules targeting the E. coli AcrAB-TolC efflux pump
(2026)
This dissertation focuses on the identification and characterization of efflux pump inhibitors targeting the main tripartite efflux pump in Escherichia coli, AcrAB-TolC. Tripartite efflux pumps are integral membrane complexes that confer antimicrobial resistance to Gram-negative bacteria by extruding antibiotics. Inhibiting efflux systems with small molecules represents a promising strategy for extending the spectrum of antibiotics, and restoring antibiotic susceptibility in multidrug-resistant bacteria. However, no efflux pump inhibitors have been approved for clinical use so far.
Two substances, LP-115 and carmofur, that represent a basis for the development of novel efflux pump inhibitors were discovered, while postulated AcrA inhibitors were shown to be non-specific binders. LP-115 was identified employing an in silico repurposing screen targeting the outer membrane factor TolC followed by microbiological validation and deconstruction of a hit compound into fragments. Binding to TolC and AcrB was confirmed using MST, and a ligand-induced destabilization of the efflux pump complex assembly was observed using dynamic light scattering. Cryo-EM provided detailed molecular insights into the binding site at the AcrA-TolC interface. Our results suggest that LP-115 is an efflux pump inhibitor with a novel mechanism of action that consists of disrupting the AcrAB-TolC efflux pump assembly. Carmofur was identified employing a microbiological repurposing screen focusing on antimicrobial potentiating effects, followed by microbiological and biophysical characterization of the interaction with the isolated efflux pump subunits using microscale thermophoresis, nano differential scanning fluorimetry, and dynamic light scattering. The synergistic activity of carmofur in combination with an AcrAB-TolC substrate was TolC-dependent and specific binding to TolC was observed. Thus, carmofur could be used as starting point for the development of novel efflux pump inhibitors.
This thesis focuses on the development and application of benchmark data sets for diverse compound classes and the differentiated assessment of docking and scoring algorithm performance using the curated sets. Various popular software, including AutoDock, AutoDock Vina, GOLD, MOE, FlexX and FITTED were assessed for two important types of compounds, which are summarized as follows.
In publication I, we investigated the fragment placement performance of molecular docking software AutoDock, AutoDock Vina, GOLD and FlexX. For this assessment we constructed LEADS-FRAG, a benchmark data set containing 93 high-quality protein-fragment complexes. GOLD with ChemPLP and AutoDock Vina performed best and generated near-native conformations (root mean square deviation <1.5 Å) for more than 50% of the data set considering the top-ranked docking pose. Taking into account all docking poses, the tested programs generated near-native conformations for up to 86% of the fragments. By rescoring with the GOLD scoring functions and PLIff, the number of near-native conformations increased up to 40% with respect to the top-rescored poses, showing that conventional small-molecule docking programs achieve a satisfactory fragment docking performance.
In manuscript 2, we examined covalently bound ligands and tested the efficiency of covalent docking options in the software programs AutoDock, GOLD, MOE and FITTED. We generated the LEADS-COV data set, containing 89 high-quality covalently bound protein-ligand complexes: 47 with a cysteine bound ligand and 42 with serine. For Cysteine GOLD with ChemPLP or ChemScore performed best and generated near-native conformations (root mean square deviation <1.5 Å) for more than 40% of the data set considering the top-ranked docking pose. Serine in comparison had better results, with over 65% top-ranked near native poses by GOLD with ChemPLP. Taking into account all generated poses values went up to over 65% for cysteine and over 80% for serine.
This thesis focuses on the role of drug repurposing in containment of an emerging disease, taking SARSCoV‐2 as a case study. The severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) emerged in2019 causing a deadly respiratory disease: COVID‐19. The increasing knowledge about SARS‐CoV‐2allowed the expansion of multiple approaches to contain the spread of infection. Eventually the rapid development of anti‐SARS‐CoV‐2 vaccines allowed a control of the pandemic. Effective antiviral pharmacological treatments are still rare and viral evolution allowed a fast adaptation and escape from available containment methods. Since the beginning of the SARS‐CoV‐2 pandemic drug repurposing was considered as a valuable source for identification of new antivirals, due to the advantage of available clinical safety data and activity profiles. To interfere with SARS‐CoV‐2 infection and to identify new antiviral compounds, key steps of the virus replication cycle and their corresponding targets were selected for assay development. The screening approaches not only identified new molecules, that can act as starting points to develop new antiviral therapies, but also revealed critical steps and pitfalls in developing assays, that will help to optimize the translation of compound effects from biochemical to cell‐based state. The study also adresses the misconception that repurposed drug cannot interfere with assay, by showing examples of compound‐reactivity through generation of reactive oxygen species, and readout interference. In addition, a drug‐combination approach for entry‐inhibitors aiming at a synergistic response is shown as an option to overcome difficulties in reaching necessary intracellular target‐doses without increasing cytotoxicity. In conclusion, this research highlights the potential of drug repurposing in antiviral drug discovery. The generated results contribute to the publicly available data on drug repurposing against SARS‐CoV‐2, which may be used for research.
Proteins are usually classified as water-soluble proteins or membrane proteins based on their cellular localization. In the course of their biogenesis, there is a significant number of initially water-soluble proteins and, after crossing into the membrane, develop their actual function as pore-forming or receptor proteins. Although the passage of the water-soluble proteins into the membrane and the associated refolding of the proteins are essential steps for the formation of the protein functions, only a few high-resolution methods exist to investigate these individual steps of the protein-membrane binding and the simultaneous development of the final active conformation in the membrane. In this study the transition of some proteins from the aqueous phase into the functional integral membrane form is examined for two cases, the human peroxisomal targeting signal 1 (PTS1) receptor hsPEX5 and the presynaptic pore-forming neu-rotoxins (LaTXs) found in the venom of Latrodectus spiders, namely α-LCT, δ-LIT and α-LTX. A vertical and horizontal artificial bilayer setup was used, enabling simultaneous and sequential high-resolution electrical and fluorescent measurements at the single-molecule level. It is presented that hsPEX5 alone harbors the ability to interact with the artificial mem-brane and generate a conductive membrane-pore. The electrophysiological results revealed for the LaTXs the essential role of calcium in stabilizing the oligomerized pore and moreover the significance of the latrotoxin-channels in cellular calcium homeostasis.
Despite success, the use of therapeutic monoclonal antibodies (mAbs) in clinical settings has been complicated by the ability of the patient’s immune system to provoke an unwanted humoral immune response against the drug, generally through the formation of anti-drug antibodies (ADAs) – termed immunogenicity. ADA onset may compromise clinical efficacy and impact safety in patients. A fundamental area of immunogenicity research is investigating mAb-derived peptides processed by dendritic cells (DCs) and presented through major histocompatibility complex (MHC) class II receptors. These mAb-derived peptides, representing potential T cell engaging epitopes, orchestrate the immunogenicity cascade by directly influencing T cell activation leading to ADA production. The MHC-II-associated peptide proteomics (MAPPs) assay is a Roche-invented methodology to identify and quantify such potential T cell epitopes. As an integrated approach during preclinical drug development, MAPPs is used alongside other in vitro, in silico, and in vivo tools to address the risk of immunogenicity. This PhD work aims to extend the applications of MAPPs for the development of a tool for personalized healthcare (PHC) in the clinic with the purpose of identifying patients with a potential risk of immunogenicity prior to treatment in order to devise an ideal treatment plan (main aim 1). Moreover, since most MAPPs studies are currently restricted to HLA-DR as the dominant MHC-II genotype due to lack of satisfactory MHC-II receptor-precipitating reagents available, an immunoprecipitation strategy using the MAPPs assay alongside the advanced epitope–prediction algorithm NetMHCIIpan was developed to accommodate MHC-II pan receptors for improved predictability of potential T cell epitopes (main aim 2). Taken together, these reformed uses of the MAPPs assay will lead to an invaluable clinical tool for immunogenicity risk assessments that support personalized healthcare.
The Molecular Mechanism of Major Histocompatibility Complex Class I Peptide Binding and Exchange
(2022)
Major histocompatibility class I (MHC-I) molecules are key to our body's immune defence against pathogens and tumors by presenting the cytosolic peptidome to cytotoxic T lymphocytes (CTLs). MHC-I/peptide complexes are presented on the cell surface, and recognition by CTLs results in a kill signal and the destruction of aberrant cells.
In this thesis, I have described the use of small molecule-assisted refolding of MHC-I proteins for generating their empty, peptide-receptive forms. I have also developed alternative methods for peptide exchange, thermostability and peptide affinity measurements, which can contribute significantly to the understanding of MHC-I selection mechanism and developing reagents for clinical applications.
The classical refolding of MHC-I in vitro has always required full-length peptides, whereas empty forms were nearly impossible to obtain. We demonstrated that both wild type and disulfide bond-stabilized MHC-I molecules can be folded with an excess of an allotype specific dipeptide. To discover specific dipeptides, I have developed a competitive enzyme-linked immunosorbent assay and used it to screen dipeptides and tripeptides that modulate MHC-I refolding. The folded dsMHC-I molecules can later be stripped of dipeptides in a washing process to generate the stable empty forms. Such empty dsMHC-I molecules can be used in rapid multimer generation.
We can also use these empty molecules to fish out peptides from tumor tissues or infected cells before characterizing them using mass spectrometry. We can also compare the peptide affinities from in-vitro assays by monitoring direct binding of the peptide to empty dsMHC-I molecules. We have successfully developed a new assay with potential clinical applications.
Taken together, in the future, we will be able to accelerate the process of neoepitope discovery and provide efficient, and robust solutions to accelerate immunotherapeutic treatments.
The amoeba Dictyostelium discoideum is a versatile and genetically tractable model organism used in research for many cellular processess. This work is focused on the biochemical study of two different processes in the amoeba.
The first part was the establishment of monitoring techniques for viral infections in the amoeba. For this purpose, two detection method should be established: a qPCR-based assay for the counting of the Acanthamoeba polyphaga mimivirus (APMV) genomes and it was observed that no net increase occurred in D. discoideum AX2 cells. APMV is taken up into the amoeba, however, the number of viral particles decreased until 6 hours post infection (hpi), indicating that APMV was not able to enter the replicative stage, possibly failing in the phagolysosomal pathway. Infection of phagolysosomal mutants indicated that lysosomal enzyme might be key in the defense against APMV.
The second part covers the potential ribosome heterogeneity during development of D. discoideum with a focus on the 2´-O-methylation (2´-O-Me) and pseudouridylation (Ψ). Both chemical modifications are introduced sequence-specifically by box C/D and box H/ACA small nucleolar RNPs (snoRNPs), respectively. Using bioinformatics and RNAseq analysis, 30 novel box C/D snoRNAs were identified, however, in silico approaches failed to deliver box H/ACA snoRNA candidates. Modified nucleotides were determined by the application of RiboMeth-seq for 2´-O-Me and HydraPsiSeq for Ψ. Several 2´-O-Me and Ψ sites were found to be substoichiometrically modified and some positions displayed dynamic modification levels during the development of the amoeba. Experiments and previous data on the U3 snoRNA implied a sudden generation of novel and potentially specialized ribosomes during development. Taken together, the presented data from D. discoideum is the first evidence for ribosome heterogeneity in the Amoebozoa supergroup, allowing to suggest that it is a common feature of all eukaryotes.
MHC (Major Histocompatibility complex) class I molecules are membrane proteins that bind intracellular peptides of eight to ten amino acids, bring them to the surface, and present them to cytotoxic T cells of the immune system. MHC class I antigen presentation is critical to protect vertebrates against pathogens. In vitro folding of MHC class I molecules, which is performed with the purpose of studying MHC class I structure and peptide binding as well as for the preparation of recombinant proteins for the detection of T cells, is assisted by adding a specific peptide to occupy the binding groove and form a stable peptide-MHC (pMHC) complex. This process is slow and must be started anew for a different peptide of interest. However, in vitro folding of class I molecules without peptide is difficult to impossible to achieve because empty class I molecules are conformationally unstable and prone to denaturation. This delays the parallel production of multiple recombinant pMHC class I molecules. Therefore, we have developed a novel method to produced empty MHC class I molecules that are stable for subsequent therapeutic applications.
Previously, our group has shown that small molecules, such as the dipeptide GM, can be used to fold class I molecules into a peptide-receptive conformation. We use these dipeptides in conjunction with a stabilized HLA-A*02:01. The additional Y84C/A139C disulfide bond links the α1/α2 helices in the F pocket region of the binding groove. In silico, molecular dynamics simulations show that this new disulfide bond, once formed, stabilizes the peptide-binding groove just like a full-length peptide. Disulfide stabilized HLA-A*02:01 can be folded with dipeptides that are subsequently removed during the purification steps. This results in empty MHC class I molecules that are stable in solution, are freeze-thaw compatible, and bind full-length peptides with rapid kinetics. Thus, the disulfide-stabilized empty class I molecules can be loaded directly
Cocoa bean fermentation encompasses the successive growth of microbial populations on the bean which results in the diffusion of microbial metabolites into the bean, as well as dramatic increases in temperature. A combination of all these events affect the bean structurally and biochemically and lead to the formation of pleasant flavour and aroma precursors.
Cocoa bean fermentation still remains an uncontrolled and spontaneously-driven process. This means that there is tremendous variety in the quality of fermented beans. A lack of markers makes it difficult to distinguish between good and bad trials of fermentation. Consequently, Western conglomerates that buy fermented beans from farmers can end up shipping large quantities of beans only to discover that they are not suitable for chocolate manufacture.
The experiments described herein were done in order to more accurately study the reactions that underlie the process of fermentation, in order to find biochemical markers defining good-quality fermentations. This was done using chemical-driven and microbial-driven model systems. Cocoa bean fermentation, assessed through bean pH, microbial dynamics and the secretions of microbial metabolites, was successfully reproduced in lab-scale quantities using starter cultures, temperature regimes and submerged incubations.
The main findings of this research were that changes in bean pH led to dramatic changes in terms of protein and polyphenol content and that an optimum pH needed to be reached in order to allow correct proteolysis and flavanol degradation for the formation of cocoa aroma and flavour. Acid influx into the bean also exhibited a preservation effect upon flavanols, which could be beneficial in preserving the health properties of cocoa. A comprehensive understanding of the factors contributing to fermentation through design of experiments also enabled the prediction of outcomes of fermentation trials.
Antibiotic resistance has become a world-wide problem as the number of resistant and multi-resistant bacteria has dramatically increased over the last years, especially for Gram-negative bacteria.
The PhD project, part of INTEGRATE, a multidisciplinary Marie Curie Educational Training Network (ETN), concerned the identification and characterization of small molecule modulators of the enzyme Caseinolytic protease proteolytic subunit (ClpP). ClpP is a serine protease and it has been proposed as an antibacterial target because of its central roles in many essential bacterial cellular processes. The main aim of this work is the identification of inhibitors directed against ClpP from Gram-negative bacteria, using E. coli as model organism.
An in vitro target-centric approach was adopted to identify novel hits by screening collections of small molecules, including diverse compounds. Selected compounds were profiled in biochemical assays and confirmed in biophysical readouts. For the most promising compounds, potential cytotoxicity against selected human cell lines was evaluated. The compound efficacy as an anti-bacterial in the presence and absence of selected stress condition and the hypothetical binding mode (using in silico approaches) were determined. In parallel, an ongoing structural biology effort is ongoing to reveal information on the interaction between selected compounds and ClpP and validate the computational models.
With this PhD thesis, further validation of the existence of phenotypic ClpP-related effects in E. coli can be found and new evidence provided for the role of ClpP as a valid target in Escherichia coli and Gram-negative bacteria antibiotic research. This thesis reports several compounds active in vitro in the low micromolar or sub-micromolar range, with an acceptable safety profile and with possible ClpP-related activity in bacteria. This study must be seen as an important starting point for further development inhibitors in follow on studies.
