Reassortment between Avian H5N1 and human influenza viruses is mainly restricted to the matrix and neuraminidase gene segments

Abstract

Highly pathogenic avian influenza H5N1 viruses have devastated the poultry industry in many countries of the eastern hemisphere. Occasionally H5N1 viruses cross the species barrier and infect humans, sometimes with a severe clinical outcome. When this happens, there is a chance of reassortment between H5N1 and human influenza viruses. To assess the potential of H5N1 viruses to reassort with contemporary human influenza viruses (H1N1, H3N2 and pandemic H1N1), we used an in vitro selection method to generate reassortant viruses, that contained the H5 hemagglutinin gene, and that have a replication advantage in vitro. We found that the neuraminidase and matrix gene segments of human influenza viruses were preferentially selected by H5 viruses. However, these H5 reassortant viruses did not show a marked increase in replication in MDCK cells and human bronchial epithelial cells. In ferrets, inoculation with a mixture of H5N1-pandemic H1N1 reassortant viruses resulted in outgrowth of reassortant H5 viruses that had incorporated the neuraminidase and matrix gene segment of pandemic 2009 H1N1. This virus was not transmitted via aerosols or respiratory droplets to naïve recipient ferrets. Altogether, these data emphasize the potential of avian H5N1 viruses to reassort with contemporary human influenza viruses. The neuraminidase and matrix gene segments of human influenza viruses showed the highest genetic compatibility with HPAI H5N1 virus.

Figure 1. Replication kinetics of H5N1-pH1N1, H5N1-H3N2 and H5N1-sH1N1 reassortant viruses in MDCK cells.

MDCK cells were inoculated with 0.01 TCID50/cell of H5N1 (black), H5 reassortant viruses harboring the NA (grey), M (red), NS (purple), NA and M (blue) and NA, M and NS (green) of pH1N1 (A), H3N2 (B) or sH1N1 (C) and supernatant samples were harvested 6, 12, 24, and 48 h later. Geometric mean titers were calculated from two independent experiments, error bars indicate standard deviations. The lower limit of detection is indicated by the dotted line.

Figure 2. Replication kinetics of H5N1-pH1N1, H5N1-H3N2 and H5N1-sH1N1 reassortant viruses in wdNHBE cells.

wdNHBE cells were inoculated with 0.02 TCID50/cell of H5N1 (black), H5N1 reassortant viruses consisting of the NA (grey), M (red), NS (purple), NA and M (blue) and NA, M and NS (green) of pH1N1 (A), H3N2 (B) or sH1N1 (C) and samples were harvested 1, 6, 12, 24, and 48 h later. Geometric mean titers were calculated from two independent experiments, error bars indicate standard deviations. The lower limit of detection is indicated by the dotted line.

Figure 3. Replication and transmission of H5N1-pH1N1 reassortant virus in ferrets.

Two ferrets were inoculated intranasally with 10^7.3 TCID₅₀ of the MDCKP1 H5N1-pH1N1 reassortant virus mixture and subsequently housed individually in transmission cages (A, B). A naïve recipient ferret was added to a cage adjacent to each transmission cage at 1dpi (A, B). Virus titers in throat (black) and nose swabs (white) of the donor ferrets (lines) and recipient ferrets (bars) were determined by endpoint titration in MDCK cells. The lower limit of detection is indicated by the dotted line.

Figure 4. Virus composition in ferrets inoculated with H5N1-pH1N1 reassortant virus during the course of infection.

Two ferrets were inoculated intranasally with 10^7.3 TCID₅₀ of the MDCKP1 H5N1-pH1N1 reassortant virus mixture. At day 1, 3, 5 and 7 dpi throat swabs were collected from both ferrets (F1 and F2). In these throat samples, as well as the MDCKP1 inoculum, the ratio of the H5N1 (red) and pH1N1 (blue) gene segments (PB2, PB1, PA, NP, NA, M and NS) was determined by pyrosequencing. ND: no detection of viral gene segments by PCR.