Recognition of cap structure by influenza B virus RNA polymerase is less dependent on the methyl residue than recognition by influenza A virus polymerase.

The cap-dependent endonuclease activity of the influenza virus RNA-dependent RNA polymerase cleaves host mRNAs to produce capped RNA fragments for primers to initiate viral mRNA synthesis. The influenza A virus (FluA) cap-dependent endonuclease preferentially recognizes the cap1 structure (m(7)GpppNm). However, little is known about the substrate specificity of the influenza B virus (FluB) endonuclease. Here, we determined the substrate specificity of the FluB polymerase using purified viral RNPs and (32)P-labeled polyribonucleotides containing a variety of cap structures (m(7)GpppGm, m(7)GpppG, and GpppG). We found that the FluA polymerase cleaves m(7)G-capped RNAs preferentially. In contrast, the FluB polymerase could efficiently cleave not only m(7)G-capped RNAs but also unmethylated GpppG-RNAs. To identify a key amino acid(s) related to the cap recognition specificity of the PB2 subunit, the transcription activity of FluB polymerases containing mutated cap-binding domains was examined by use of a minireplicon assay system. In the case of FluA PB2, Phe323, His357, and Phe404, which stack the m(7)GTP, and Glu361 and Lys376, which make hydrogen bonds with a guanine base, were essential for the transcription activity. In contrast, in the case of FluB PB2, the stacking interaction of Trp359 with a guanine base and putative hydrogen bonds using Gln325 and Glu363 were enough for the transcription activity. Taking these results together with the result for the cap-binding activity, we propose that the cap recognition pocket of FluB PB2 does not have the specificity for m(7)G-cap structures and thus is more flexible to accept various cap structures than FluA PB2.