N⁶-methyladenosine regulates Influenza A virus mRNA stability yet is rarely found on genomic RNA

Previous studies have found widespread N⁶-methyladenosine (m⁶A methylation) on all forms of Influenza A virus (IAV) RNA, with m⁶A found critical for viral replication, pathogenicity as well as viral RNA packaging. Here we applied the latest quantitative technologies to revisit the methylation landscape on the anti-sense genomic RNA of IAV. Unexpectedly, upon Ultra-Performance Liquid Chromatography-Tandem Mass Spectrometry (UPLC-MS/MS) analysis of IAV virion -extracted genomic RNA, we detected very little m⁶A regardless of production from human cells or chicken eggs. Concordantly, Nanopore direct RNA sequencing also detected an overall low occurrence and stoichiometry (generally <5%) of m⁶A across all viral genomic RNA segments, compared with abundant m⁶A sites on viral mRNAs at ~20-30% m⁶A. Cross validation with glyoxal- and nitrite-mediated deamination of unmethylated adenosines (GLORI) confirmed multiple m⁶A sites on viral mRNA yet very few m⁶A on the genomic RNA. This paucity of m⁶A on genomic RNA makes it unlikely that m⁶A contributes to viral RNA packaging. Knockdown or pharmacological inhibition of the m⁶A methyltransferase METTL3 as well as the reader protein YTHDF2 both reduced viral mRNA levels and infectious viral particle production, with YTHDF2 promoting viral mRNA stability. Thus, the presence of m⁶A on IAV transcripts is indeed proviral, yet it is the mRNAs instead of genomic RNAs that are methylated at functionally relevant levels. Lastly, we provide proof of concept that a METTL3 small molecule inhibitor can be antiviral, and propose that m⁶A-targeted antivirals would mainly impact the intracellular gene expression phase of IAV replication.