Antiviral mechanisms and preclinical evaluation of amantadine analogs that continue to inhibit influenza A viruses with M2S31N-based drug resistance

Abstract

To better manage seasonal and pandemic influenza infections, new drugs are needed with enhanced activity against amantadine- and rimantadine-resistant influenza A virus (IAV) strains containing the S31N variant of the viral M2 ion channel (M2^S31N). Here we tested 36 amantadine analogs against a panel of viruses containing either M2^S31N or the parental, M2 S31 wild-type variant (M2^WT). We found that several analogs, primarily those with sizeable lipophilic adducts, inhibited up to three M2^S31N-containing viruses with activities at least 5-fold lower than rimantadine, without inhibiting M2^S31N proton currents or modulating endosomal pH. While M2^WT viruses in passaging studies rapidly gained resistance to these analogs through the established M2 mutations V27A and/or A30T, resistance development was markedly slower for M2^S31N viruses and did not associate with additional M2 mutations. Instead, a subset of analogs, exemplified by 2-propyl-2-adamantanamine (38), but not 2-(1-adamantyl)piperidine (26), spiro[adamantane-2,2′-pyrrolidine] (49), or spiro[adamantane-2,2′-piperidine] (60), inhibited cellular entry of infectious IAV following pre-treatment and/or H1N1 pseudovirus entry. Conversely, an overlapping subset of the most lipophilic analogs including compounds 26, 49, 60, and others, disrupted viral M2-M1 protein colocalization required for intracellular viral assembly and budding. Finally, a pilot toxicity study in mice demonstrated that 38 and 49 were tolerated at 30 mg/kg. Together, these results indicate that amantadine analogs act on multiple, complementary mechanisms to inhibit replication of M2^S31N viruses.