Belfort Group

Biomolecular Separations Research Group
Howard P. Isermann Department of Chemical and Biological Engineering
and Center for Biotechnology and Interdisciplinary Studies
Rensselaer Polytechnic Institute
Troy, NY 12180, USA


Antimicrobial Peptide Mechanisms

Antibiotic resistance is increasing at an alarming rate. Life depends on the integrity of biological membranes and their ability to maintain function. Disrupting bacterial but not mammalian membrane integrity with peptides is an alternate strategy to destroy toxic bacteria. Mechanisms of AMP disruption and their effect on bacterial membranes are summarized in Fig. 1. We contend that potent, stable, amphipathic AMPs together with a fundamental understanding of their mechanism of cell disruption are urgently needed. Low stability, inadequate bacterial killing, high manufacturing cost and unclear disruption mechanisms limit the enormous potential of AMPs as novel antibiotics for therapeutic applications. In addition, we aspire, not only to learn Nature's rules for potency and stability of AMPs, but with modern protein engineering and rational methods, we aim to improve and surpass the performance of native AMPs.


The goal of our research is to determine if four antimicrobial peptides (AMPs) with different secondary structure in solution interact differently with mimics of bacterial and mammalian outer membranes. Indolicidin (random/extended), protegrin-1 (β-sheet), α-defensin-1 (β-sheet), and magainin-2 (α-helix) were selected. The interactions of these AMPs with mimics of bacterial and mammalian supported lipid bilayers (SLB) were assessed by measuring changes in frequency and dissipation (i.e. rigidity) of the SLBs with time. Temporal measurements were made with a Quartz Crystal Microbalance with Dissipation (QCM-D) and analyzed with Voigt model to obtain changes in lipid bilayer surface mass density, shear viscosity and shear modulus with time (Fig. 2A & B). Protegrin-1 with its β-sheet solution structure exhibited the highest selectivity for the bacterial over the mammalian membrane and a the strongest drop in shear viscosity and shear modulus(Fig. 2A & B). The random/extended solution structure of indolicidin incurred the second greatest mass increase for the bacterial SLB and the highest mass density increase for the mammalian SLB. Both α-defensin-1 and magainin-2 with their β-sheet and α-helix solution structures, respectively, showed little propensity for binding to either bacterial or mammalian mimic membranes. None of the AMPs were able to remove mass from the lipid bilayers. Protegrin-1 also showed the highest initial efflux (slope) of calcein leakageand hence pore formation, while α-defensin-1 had the lowest leakage (Fig. 2C).

We collaborate with Kathleen McDonough, Wadsworth Center Albany,NY and Myriam Cotten, Hamilton College.