1 National Food Institute, Technical University of Denmark2 National Institute of Aquatic Resources, Technical University of Denmark3 Bacterial Ecophysiology and Biotechnology, Department of Biotechnology and Biomedicine, Technical University of Denmark
Antibiotics have been an effective weapon against bacterial infections for over 50 years. However, bacterial resistance towards conventional antibiotics has increased considerably within the last decades and the number of antibacterial agents available for treating complicated bacterial infections is becoming increasingly limited. In the search for alternatives therapies, antimicrobial peptides (AMPs) have received considerable attention since they target the bacterial Achilles’ heel i.e. their distinct membrane structure. These host defence molecules are ubiquitous in nature by forming part of the immune system among all classes of life. Several of these compounds have therefore been characterised and developed into future antibacterials. Furthermore, in an attempt to improve the antibacterial activity, synthetic analogues i.e. peptidomimetics have been designed based on the structural properties of natural AMPs. The purpose of this PhD study was to establish the potential correlation between structure and antibacterial activity for a series of a-peptides/ß-peptoid peptidomimetics and additionally to determine if mechanistic differences could explain observed variations in activity. We determined the activity of the peptidomimetics against a range food borne and nosocomial pathogenic bacteria. These structure-activity studies demonstrated that peptide length was important for high antibacterial activity since analogues with a length shorter than 12 residues were virtually inactive. In the present design, with a 1:1 ratio between cationic a-amino acids and hydrophobic ß-peptoids, amino acid composition and chirality in the ß-peptoid unit only had a minor influence on antibacterial activity. By using an ATP leakage assay we determined that the mechanism of action of the chimeras was permeabilization or disruption of the bacterial cell membrane. The resulting changes to the cell surface were visualised with Scanning Electron Microscopy (SEM). Importantly, our leakage studies were performed with viable bacterial cells and using a concentration that was close to the Minimum Inhibitory Concentration (MIC). The findings show that all of the chimeras included in the study have a similar mechanism of action that was independent on bacterial species. However, the study showed that the detailed interaction with the cell membrane may be different, since there were large variations in the amount of leaked ATP and subsequent loss of viability. A series of three peptides differing only in length all caused ATP leakage but only the longest of the three caused complete depletion of intracellular ATP, which correlated with a substantial loss in the number of viable cells. In a continuous selection protocol encompassing 500 generations, 10 out of 10 lineages of Escherichia coli developed resistance towards the chimera they had been exposed to. This was the first time resistance was successfully developed towards peptidomimetics, though several studies have reported resistance towards AMPs. Resistance was specific to compounds within the peptidomimetics library, since we were unable to demonstrate cross-resistance to other AMPs. We sequenced the entire genome of six highly resistant isolates from two separate lineages, and identified a single-nucleotide-polymorphism (SNP) in the gene encoding the MltD protein. This protein functions in the reorganization of the peptidoglycan layer, and we consider it likely that a change in this protein is the cause of resistance, since the SNP was found exclusively in isolates with high levels of resistance. Conversely, these resistant isolates displayed increased sensitivity towards human blood plasma possibly due to immune effector compounds present in this. The addition of 50 % blood plasma also increased the activity of the chimeras against wild type bacteria by up to 32 times. This effect was abolished by heat-treatment, which is a method known to inactivate the complement system. The findings in this thesis have elucidated how central structural determinants influence antibacterial activity. Peptidomimetics can be regarded as promising future antibiotics since the possibility to optimize their properties through structural modification allows for continuous variation. This thesis concludes that antibacterial activity can be improved further and that in the future resistance may be circumvented by optimizing the existing scaffold.