A distinct lineage of influenza A virus from bats

Abstract

Influenza A virus reservoirs in animals have provided novel genetic elements leading to the emergence of global pandemics in humans. Most influenza A viruses circulate in waterfowl, but those that infect mammalian hosts are thought to pose the greatest risk for zoonotic spread to humans and the generation of pandemic or panzootic viruses. We have identified an influenza A virus from little yellow-shouldered bats captured at two locations in Guatemala. It is significantly divergent from known influenza A viruses. The HA of the bat virus was estimated to have diverged at roughly the same time as the known subtypes of HA and was designated as H17. The neuraminidase (NA) gene is highly divergent from all known influenza NAs, and the internal genes from the bat virus diverged from those of known influenza A viruses before the estimated divergence of the known influenza A internal gene lineages. Attempts to propagate this virus in cell cultures and chicken embryos were unsuccessful, suggesting distinct requirements compared with known influenza viruses. Despite its divergence from known influenza A viruses, the bat virus is compatible for genetic exchange with human influenza viruses in human cells, suggesting the potential capability for reassortment and contributions to new pandemic or panzootic influenza A viruses.

Fig. 1.

Locations of bat-sampling sites in Guatemala. Geographic locations in Guatemala in which bats were captured are indicated with red dots. Locations of pan-Flu RT-PCR–positive bats are denoted by arrows. See SI Appendix, Table S1 for additional information.

Fig. 2.

Structural and functional elements of the A/bat/Guat/09 HA and NA-like proteins. HA and NA proteins from A/bat/Guat/09 and those of known influenza A viruses were aligned and compared. (A, Upper) Amino acid similarity plot of HA from A/South Carolina/1918(H1N1), A/Panama/2007/1999(H3N2), and A/bat/Guat/09 as shown in the legend. (A, Lower) Schematic diagram of aligned HA from influenza A/South Carolina/1918(H1N1) (upper half of the horizontal bar) and A/bat/Guat/09 (lower half), showing major structural sections: signal peptide (SP) in solid green, HA1 in yellow, fusion peptide in red, HA2 in pink, and transmembrane domain (TM) in gray; putative glycosylation sites (cyan lollipops); conserved disulphide bonds (blue lines); and residues involved in sialic acid binding (green pin). Receptor-binding residues (green pins) are shown for A/bat/Guat/09 if the residues are found in known influenza A viruses rather than being unique to A/bat/Guat/09. Gaps in the alignment are indicated with red triangles. (B, Upper) Amino acid similarity plot of NA from A/bat/Guat/09, A/Tokyo/3/1967(H2N2) and B/Lee/40 as shown in the legend. (B, Lower) Schematic diagram of aligned NA from A/Tokyo/3/1967(H2N2) and A/bat/Guat/09. Labels are as in A, except that green pins represent residues involved in sialidase catalytic activity.

Fig. 3.

Evolution of the A/bat/Guat/09 genome. Phylogenetic trees of the eight gene segments of the bat virus genomes (represented by a red dot and red branch) were inferred by the maximum-likelihood method using the dataset described in SI Appendix, Table S7. Bootstrap values are shown above branches to the left of major nodes denoting influenza A (shaded in light blue), influenza B (indicated in purple), and A/bat/Guat/09 (indicated by a red dot) divergences. (Scale bars, number of expected nucleotide substitutions per site.) Trees are shown for (A) HA, (B) NA, (C) PB2, (D) PB1, (E) PA, (F) NP, (G) M, and (H) NS segments.

Fig. 4.

RNA-dependent RNA polymerase activity of A/bat/Guat/09 ribonucleoprotein (RNP) complex proteins (PB2, PB1, PA, and NP). A549 human lung cells were transfected with pPol1-A/bat-NS.NCR-Renilla or pPol1-A/WSN-NS.NCR-Renilla and pSV40-Luc reporter plasmids together with plasmids expressing PB2, PB1, PA, and NP from either A/WSN/33 or A/bat/Guat/09 viruses (4RNP; solid bars) or without the PB1 expression plasmid (3 RNP ctrl; open bars). Values shown represent the activities of each RNP and reporter relative to that of WSN virus (100%). Error bars indicate 95% confidence intervals. Experiments were performed three times independently.