archives@tulane.edu / Despite great progress in the treatment viral illnesses, epidemics of emerging infectious diseases often have no specific treatments, leaving most patients to be treated with only supportive care. Broad-spectrum antiviral drugs could help ease this burden especially in situations where diagnostics are limited. Recently, many antiviral peptides have been discovered to inhibit a wide array of viruses using the Wimley-White interfacial hydrophobicity scale to predict inhibitory peptides, some of which were found to possess broad-spectrum activity. These broad-spectrum peptides share some common characteristics: they are moderately hydrophobic and have a propensity to disrupt membranes. We hypothesize that there are general molecular characteristics that these peptides possess, such as a propensity for membrane interaction, that drive broad-spectrum antiviral activity, which we can utilize for the design of novel potent antiviral peptides. To test this hypothesis, we first selected membrane-active peptides with no previously known antiviral activity. We screened these peptides against multiple diverse viruses and found that a majority possess antiviral activity with little cytotoxicity. We next explored the mechanism of action and found that peptides bind virus and cause rapid aggregation. We then used this information to guide the design of novel gain-of-function antiviral peptides. We sequentially designed peptides to test specific hypotheses about sequence motifs that may alter antiviral activity and/or cytotoxicity. Through our screenings we uncovered sequence motifs that led to greater antiviral activity, lower cytotoxicity, and potent broad-spectrum activity. More optimization is needed to turn these peptides into viable drug candidates, but we believe our findings identified important design aspects of antiviral peptides that could lead to the creation of potent broad-spectrum antiviral therapeutics. / 1 / Andrew Hoffmann