Pathogenicity And Transmissibility Of The Novel Reassortant H5 Subtype Highly Pathogenic Avian Influenza Virus

Historically, several devastating influenza pandemics have been caused by viruses generated through genetic reassortment such as the H2N2 influenza virus responsible for the pandemic of 1957, which was generated by the prevailing human influenza A virus (H1N1) acquiring the HA (H2), NA(N2), and PB1 genes from an avian virus. Similarly, H3N2 virus responsible for the 1968 pandemic arose through the acquisition of novel HA (H3) and PB1 genes of avian origin in H2N2 virus while the 2009 pandemic H1N1 virus was a quadruple reassortant that contained PB2 and PA genes of North American avian virus origin. PB1 gene of human H3N2 virus origin. HA (H1), NP, and NS genes of classical swine virus origin, and NA (N1) and M genes of Eurasian avian-like swine virus origin. Moreover, the novel reassortant human-infecting H7N9 avian influenza virus which was initially reported in the Yangtze River Delta of China in March 2013 has also been demonstrated to be a natural reassortant of three avian viruses:H7N3, H10N9 and H9N2. Its two surface genes derive from H7N3 and H10N9 viruses respectively, and the six internal genes all derive from endemic H9N2 avian viruses in poultry. In China, H5 subtype highly pathogenic avian influenza virus (HPAIV) was firstly isolated in 1996 as represented by A/goose/Guangdong/1/96(H5N1), and has continued to evolve into over 10 distinct phylogenetic clades including different subclades based on the HA gene. It is well known that H5N1 HPAIV poses a serious pandemic threat due to its occasional infection and high mortality in human beings, and sporadic human infections continue to occur in countries where H5N1 viruses are endemic. Although the current H5N1 strains have not yet acquired efficient human-to-human transmissibility, various reassortant viruses have continuously been detected in different domestic poultry more recently, with the HA gene derived from the A/goose/Guangdong/1/96-like (Gs/GD-like) lineage and the other genes from other subtypes such as H9N2. Therefore, understanding the pathogenicity and transmissibility of the novel reassortant H5 subtype HPAIV will facilitate designing strategy useful for the prevention and control of influenza endemics and pandemics.It has been reported previously that the H5N2 HPAIV A/chicken/Hebei/1102/2010 (HB10) is a natural reassortant between circulating Clade 7.2 H5N1 and endemic H9N2 influenza viruses. In addition, the internal genes are all highly homologous to the novel reassortant avian-origin influenza A (H7N9) virus. We surmised that the novel reassortant H5N2 may become adaptive in human beings similar to the H7N9 virus. To evaluate the potential of its interspecies transmission, the wild-type HB10 was adapted in mice through serial lung passages. Increased virulence was detectable in 5 sequential lung passages in mice and a highly virulent mouse-adapted strain (HB10-MA) with a 50% mouse lethal dose of 1025 was obtained in 15 passages. The virulence and the replication efficiency of HB10-MA in mice were significantly higher than those of HB10 while HB10-MA grew faster and to significantly higher titers than HB10 in MDCK and A549 cells. Only five amino acid mutations in four viral proteins (HA-S227N, PB2-Q591K, PB2-D701N, PA-1554V and NP-R351K) of HB10-MA virus were found when compared with those of the parent HB10, indicating that they may be responsible for the adaptation of the novel reassortant H5N2 avian influenza virus in mice with increased virulence and replication efficiency.We further investigated the molecular basis for the enhanced virulence of HB10-MA in mice. By generating a series of reassortants between the two viruses and evaluating their virulence, we found that PB2 and PA genes contribute to the high virulence of HB10-MA in mice while PB2 gene played a major role. Considering that there are only two amino acid differences in the PB2 gene between HB10 and HB10-MA viruses, site-directed mutants were generated to pinpoint which amino acid affects virulence. The results showed that both residues of 59IK and 701N in PB2 gene contribute to the high virulence of HB10-MA in mice, and they have synergistic effect on enhancing the virulence, virus replication, and polymerase activity. Our results collectively suggest that paired mutations of the Q591K and D701N in PB2 play a key role in dictating the virulence of the novel reassortant H5N2 virus in mice. Moreover, as mutation E627K in the PB2 protein has previously been identified as high-virulence-related determinant of avian and pandemic influenza viruses in mammals, PB2 mutant viruses carrying various residue patterns at 591,627 and 701 in the background of HB10 were rescued to investigate the contribution of these three sites to the virulence of the natural reassortant H5M2 HPAIV in mammals. We found that these amino acid substitutions resulted in increased mouse pathogenicity and better mammalian adaptation in terms of virus replication and polymerase activity in vitro. Moreover, D701N paired with Q591K or E627K mutation potentiated virulence of HB10 virus, compared to those with single mutations. Unexpectedly,627K. paired with 591K did not potentiate but rather weakened its virulence. Therefore, our study emphasizes the crucial role of PB2 gene in the host adaptation of avian influenza virus in mammals and may provide helpful insights into the pathogenic potential of the novel reassortant H5N2 viruses in humans.Apart from the novel H5N2 reasortants with their HA genes belonging to Clade 7.2, natural reassortant viruses with various NA subtypes (H5N1, H5N2, H5N5, H5N6 and H5N8) but retained the HA genes from Clade 2.3.4 variant H5N1 HPAIV have continuously been detected in different domestic poultry in China more recently. For example, the novel reassortant H5N6 avian influenza virus has resulted several outbreaks in poultry in the Southeast Asian and caused human death in China, and the H5N8 virus has invaded into at least 11 countries in wild birds and poultry. As is generally accepted that haemagglutinin-receptor-binding preference to a-2,6-linked sialylated glycans is the initial key step for a novel influenza-virus-causing pandemic, we first evaluated the receptor binding property of these reassortant Clade 2.3.4 variant H5 HPAIVs. The results showed that these viruses bound to both avian-type (a-2,3) and human-type (a-2,6) receptors. Although receptor binding preference is a major factor determining host range and tissue tropism, we do not know whether the alteration in receptor binding properties is sufficient to enable the virus to transmit from person to person efficiently. We further used guinea pigs as a mammalian model to examine the replication and transmission of these reassortant viruses. We found that two of these Clade 2.3.4 variant viruses,031(H5N5) and GS/EC/1112/11(H5N2), not only replicated well but also transmitted efficiently in guinea pigs. These findings emphasize that continued circulation of these viruses may pose health threats for humans. Therefore, we need to intensify our effort to detect such virus as early as its first emergence.Alignment of the HA gene sequences from Clade 2.3.4 variant viruses with that of a classic Clade 2.3.4 virus HD/05 revealed 8 amino acid differences, and these residues were chosen for further evaluation. We conducted gene mutations analysis through reverse genetics to determine which mutations contributed to the receptor binding property of these reassortant Clade 2.3.4 variant viruses. We found that T160A mutation resulting in the lack of an oligosaccharide side chain at 158-160 of HA, was responsible for the HA to recognize the human-type receptor. We further investigated whether this amino acid affect the replication and transmission of Clade 2.3.4 H5 viruses in guinea pigs, and the results showed that the mutant virus not only replicated well but also transmitted efficiently in guinea pigs. We also confirmed the previous finding that elimination of 158N glycosylation site enhanced viral virulence and systemic spread in mice. In addition, the mutant lacking the glycosylation site at position 158 showed slightly higher titers than wild-type virus in both MDCK and A549 cells. Our results collectively suggest that the absence of a potential N-linked glycosylation site at HA amino acid positions 158-160 may serve as an important molecular marker for assessing the pandemic potential of Clade 2.3.4 H5 variant isolates. Moreover, mutants with high avidity for both human and avian receptors may be intermediates in the evolution of Clade 2.3.4 variant H5 viruses that could infect both humans and poultry. Therefore, continuing surveillance of such viruses could serve as an early warning system for a potential pandemic.In summary, we demonstrated that mutations in PB2 protein were critical for the adaptation and virulence of the novel reassortant H5N2 virus in mice by using a highly virulent mouse-adapted virus while we found that two amino acids, PB2 Q591K and D701N an important role synergistically in the high virulence of the HB10-MA virus. Moreover, the absence of a potential N-linked glycosylation site at HA amino acid positions 158-160 is critical for the novel reassortant Clade 2.3.4 variant H5 viruses to bind to both a-2,3 and a-2,6 glycans. Therefore, the presence or absence of 158N glycosylation site may serve as an important molecular marker for assessing the pandemic potential of the recent Clade 2.3.4 variant H5 isolates bearing various NA subtypes. These findings emphasize that the importance of surveillance of H5 subtype influenza viruses which may provide crucial data for formulating strategies to effectively prevent and control influenza endemics and pandemics.