Delivery, design, and mechanism of antimicrobial peptides

Abstract

Each year, 2 million people contract hospital-acquired bacterial infections, which causes the death of 100,000 patients and costs the US healthcare system over $21 billion. These infections have become dangerously resistant to our existing line of antibiotics and are rapidly spreading outside of hospitals and into communities. As molecular targets to develop new antibiotics are becoming exhausted, clinicians and scientist are concerned that antibiotic resistant infections will wipe out most of the major health benefits acquired over the last century. The work described in this thesis develops new antimicrobials strategies against bacterial infections, focusing on antimicrobial peptides (AmPs). We first delivered genes inducing the toxic expression of AmPs and other lytic agents directly into bacteria using re-engineered bacteriophages. Expression of these lytic agents in lysogenic bacteriophages resulted in bactericidal activity, and demonstrated, for the first time, a long-term cidal effect for over 20 hours. We then enhanced the efficacy of our approach by expressing the same agents in lytic bacteriophage, which resulted in complete suppression of the bacterial culture and prevented bacterial regrowth and resistance to bacteriophages. Since a large fraction of medical infections originates at the surface of implantable devices, we developed film coatings that release active AmPs to cover these surfaces and prevent bacterial colonization. We incorporated AmPs in layer-by-layer films and demonstrated that the kinetics of AmP release can be adjusted. These released AmPs still actively prevented bacterial growth and remained non-toxic towards mammalian cells. While natural AmPs have broad activity against pathogens, they are not optimized for a specific antimicrobial function or bacterial target. Thus, researchers have tried for decades to design highly active and specific de novo AmPs. One approach is to design new peptides using conserved motifs identified from the amino acid sequence of natural AmPs. We improved this approach by measuring the antimicrobial activity of a large database of natural AmPs and incorporating this activity information in the design algorithm. This strategy improved the success rate of designing de novo peptides from 45% to 73% and increased the antimicrobial strength of the designed peptides. Finally, we developed new potentiating strategies by studying the mode-of-action of the family of ponericin AmPs. First, we measured their cidal behavior and differentiated bactericidal ponericins from bacteriostatic ones. Using a modified AFM and a microfluidic device, we observed that the action of AmPs led to cellular death through the corrugation of bacterial, while subpopulation of cells resisted the action of the AmPs longer than others. Focusing on the ponericin G1 AmP, we correlated these visual observations with various membrane stress sensing mechanisms. We concluded that bacteria's ability to develop resistance to ponericin G1 requires the sensing and repair of misfolded membrane proteins via the CpxAR system, as well as DNA repair via induction of the SOS response by RecA. Using microarrarys, we showed that ponericin G1 targets tRNA synthetases in the ribosome. Finally, we demonstrated 99.999% killing of antibiotic resistant bacteria by potentiating ponericin G1 with the ribosomal antibiotic kanamycin, whereas no killing is observed when these two agents are applied independently. untreta The PhDCEP capstone requirement finalizes the work of this thesis by analyzing market entry and expansion strategies for an antimicrobial company commercializing genetically engineered bacteriophages. In conclusion, this thesis establishes new advances in the delivery, the design and the potentiation of AmPs in order to eradicate resilient bacterial infections.