De Novo Discovery of Bicyclic Competitive Inhibitors of Zika Virus Protease from Peptide–Bismuth Phage Display Screening

No specific vaccines or therapeutics are currently available for the prevention or treatment of Zika virus infections. The viral protease NS2B-NS3 is essential for the replication of Zika and other orthoflaviviruses, making it a target for antiviral drug development. Traditional discovery of competitive inhibitors has relied on substrate recognition sequences, typically yielding multi-basic peptides. Here, we present a de novo strategy for identifying competitive inhibitors using peptide phage display in combination with Bi(III)-mediated in situ formation of bicyclic peptides. In an initial screening, phages displaying a library of randomized peptide-bismuth bicycles were eluted by interrupting the phage-target interactions at low pH. This approach yielded a small number of peptides biased toward the active site, characterized by dibasic motifs, but only one low-ranking sequence showed modest inhibitory activity. To enhance specificity, a second screening campaign employed competitive phage elution using the dibasic boronate inhibitor CN-714 that covalently binds to the catalytically active serine residue S135 of NS2B-NS3. This strategy enriched a larger pool of competitive inhibitors sharing the characteristic dibasic substrate recognition motif. The most potent peptide-bismuth bicycle identified featured a completely novel sequence, exhibited an inhibition constant of 3.9 µM and displayed remarkable proteolytic stability over 24 hours.