Influenza A virus PB1-F2 gene in recent Taiwanese isolates

Abstract

Influenza A virus contains eight RNA segments and encodes 10 viral proteins. However, an 11th protein, called PB1-F2, was found in A/Puerto Rico/8/34 (H1N1). This novel protein is translated from an alternative open reading frame (ORF) in the PB1 gene. We analyzed the PB1 gene of 42 recent influenza A isolates in Taiwan, including 24 H1N1 and 18 H3N2 strains. One H1N1 isolate and 17 H3N2 isolates contained the entire PB1-F2 ORF of 90 residues, three amino acids (aa) longer than the PB1-F2 of A/Puerto Rico/8/34 at the C terminal. The one remaining H3N2 isolate encoded a truncated PB1-F2 with 79 residues. The other 23 H1N1 isolates contained a truncated PB1-F2 of 57 aa. Phylogenetic analysis of both the HA and the PB1 genes showed that they shared similar clustering of these Taiwanese isolates, suggesting that no obvious reassortment occurred between the two genomic segments.

Figure 1

Number of isolated influenza A viruses in Clinical Virology Laboratory, Chang Gung Memorial Hospital, a contract laboratory of the Center for Disease Control and Prevention—Taiwan.

Figure 2

Alignment of putative PB1-F2 amino acid sequences of 24 Taiwanese H1N1 strains and 10 H1N1 reference strains. Most strains, including 23 Taiwanese strains and 5 reference strains, contained a truncated open reading frame (ORF) 57 residues long. One reference strain, A/Wisconsin/3523/88, had the ORF truncated at 11 residues. PB1-F2 of A/Puerto Rico/8/34 with 87 residues is placed on top of the alignment. One Taiwanese strain and three reference strains each contained a complete PB1-F2 of 90 residues. The program AlignX in VectorNTI Suite produced the alignment.

Figure 3

Alignment of putative PB1-F2 amino acid sequences of 18 Taiwanese H3N2 strains and 17 H3N2 reference strains. PB1-F2 of A/Puerto Rico/8/34 (H1N1) with 87 residues is laid over the alignment for reference. Most strains contained a PB1-F2 ORF 90 residues long. One Taiwanese strain, A/Taiwan/1748/97, encoded a truncated open reading frame (ORF) with 79 residues, and one reference strain, A/Shiga/25/97, encoded a 87-residue product as in A/Puerto Rico/8/34 (H1N1). The program AlignX in VectorNTI Suite produced the alignment.

Figure 4

Phylogenetic tree of influenza A viruses for their PB1 gene nucleotide sequences. Apart from the 42 Taiwanese isolates obtained in this study, 27 reference strains were included; these were selected on the basis of an extensive search of all human H1N1 and H3N2 influenza viruses from GenBank, whose PB1 sequences were shown to be able to be translated into the putative PB1-F2 gene. The tree was rooted with B/Lee/40. All strains were separated into two groups according to their subtypes. Sequence analysis was conducted using the software Lasergene, Clustal W and PHYLIP with 1,000 replicates. All sequences are 300-nt long from genomic position 104 to 403 based on the PB1 gene of A/Puerto Rico/8/34, which covers at least the entire PB1-F2.

Figure 5

Phylogenetic tree of H1N1 influenza A viruses for HA nucleotide sequences. Five recent vaccine strains were included along with the 24 Taiwanese strains and six reference strains considered previously in Figure 4. The tree was rooted with A/Puerto Rico/8/34. Sequence analysis was conducted using the software Lasergene, Clustal W, and PHYLIP with 1,000 replicates. Sequences contained 561 to 564 nt bases from positions 120 to 683, based on A/Puerto Rico/8/34.

Figure 6

Phylogenetic tree of influenza A H3N2 viruses for HA nucleotide sequences. Six recent vaccine strains were included along with the 17 Taiwanese strains and 13 reference strains shown in Figure 4. The tree was rooted with A/England/939/69. Sequence analysis was conducted by using the software Lasergene, Clustal W, and PHYLIP with 1,000 replicates. All sequences are 844-nt long from positions 187 to 1031, based on A/Hong Kong/1/68.