Evolution of H5-Type Avian Influenza A Virus Towards Mammalian Tropism in Egypt, 2014 to 2015

Abstract

Highly pathogenic avian influenza viruses (HPAIV) of the H5-subtype have circulated continuously in Egypt since 2006, resulting in numerous poultry outbreaks and considerable sporadic human infections. The extensive circulation and wide spread of these viruses in domestic poultry have resulted in various evolutionary changes with a dramatic impact on viral transmission ability to contact mammals including humans. The transmitted viruses are either (1) adapted well enough in their avian hosts to readily infect mammals, or (2) adapted in the new mammalian hosts to improve their fitness. In both cases, avian influenza viruses (AIVs) acquire various host-specific adaptations. These adaptive variations are not all well-known or thoroughly characterized. In this study, a phylogenetic algorithm based on the informational spectrum method, designated hereafter as ISM, was applied to analyze the affinity of H5-type HA proteins of Egyptian AIV isolates (2006-2015) towards human-type cell receptors. To characterize AIV H5-HA proteins displaying high ISM values reflecting an increased tendency of the HA towards human-type receptors, recombinant IV expressing monobasic, low pathogenic (LP) H5-HA versions in the background of the human influenza virus A/PR/8/1934(H1N1) (LP 7+1), were generated. These viruses were compared with a LP 7+1 expressing a monobasic H5-HA from a human origin virus isolate (human LP-7271), for their receptor binding specificity (ISM), in vitro replication efficiency and in vivo pathogenicity in mammals. Interestingly, using ISM analysis, we identified a LP 7+1 virus (LP-S10739C) expressing the monobasic H5-HA of AIV A/Chicken/Egypt/S10739C/2015(H5N1) that showed high affinity towards human-type receptors. This in silico prediction was reflected by a higher in vitro replication efficiency in mammalian cell cultures and a higher virulence in mice as compared with LP-7271. Sequence comparison between the LP-S10739C and the LP-7271 H5-HA, revealed distinct amino acid changes. Their contribution to the increased mammalian receptor propensity of LP-S10739C demands further investigation to better deduce the molecular determinant behind the reported high morbidity of 2014 to 2015 HPAI H5N1 virus in humans in Egypt. This study provides insights into the evolution of Egyptian H5 HPAIVs and highlights the need to identify the viral evolution in order to recognize emerging AIV with the potential to threaten human and animal populations.

Keywords: ISM analysis; evolution; human-type receptor.

Figure 1

ISM phylogenetic analysis of the avian 2014 to 2015 H5-HA viruses´ affinity to mammals. On the basis of the ISM analysis, seven viruses were selected from different clusters to be tested in vitro, (S10738B, D10551C, D10552B, Q10920C, A10540A, S10739C, and A/du/Eg/4/2015).

Figure 2

Receptor binding specificity of avian H5-HA viruses to mammalian- and avian-type receptors. The PR8 viruses expressing monobasic HA from clade 2.2.1.2 H5N1 strains (LP-S10738B, LP-D10551C, LP-D10552B, LP-Q10920C, LP-S10739C, LP-A10540A, and LP-A/Du/Eg/4/2015); clade 2.2.1 (LP-2006), and clade 2.2.1.1 (LP-Q1995D, LP-M2583A, and LP-M7217B), were compared to LP-7271 (clade 2.2.1.2, H5 HA of human origin) using solid-phase receptor binding assay.

Figure 3

Replication kinetics of LP H5-HA viruses in MDCK-SIAT1 cells. The cells were infected with the tested viruses at different time points (6, 12, 24, and 36 h) at a MOI of 0.001. Unlike other PR8 expressing the 2014 to 2015 H5 HA, the LP-S10739C replicated significantly higher than LP-7271 at different time points 12, 24, and 36 h p.i. Error bars reflect standard deviation (SD) of three independent experiments. Statistical analysis was performed using repeated measures ANOVA, followed by Bonferroni post-hoc test. The significant differences are indicated (* = p < 0.05, ** = p < 0.01, *** = p < 0.001, and nonsignificant = ns).

Figure 4

Replication kinetics of LP H5-HA viruses in A549 cells. The cells were infected with the tested viruses at different time points (6, 12, 24, and 36 h) at a MOI of 0.001. All 2014 to 2015 viruses were highly replicating than human LP-7271 strain especially at 24 h p.i. the replication efficiencies of LP-A10540A, LP-D10551C, LP-D10552B, and LP-A/du/Eg/4/2015 were significantly higher than LP-7271 at 24 h or 36 h p.i. Error bars reflect standard deviation (SD) of three independent experiments. Statistical analysis was performed using repeated measures ANOVA, followed by Bonferroni post-hoc test. The significant differences are indicated (* = p < 0.05, ** = p < 0.01, *** = p < 0.001, and nonsignificant = ns).

Figure 5

Virulence of LP AIVs in mice. Each mouse was infected with 30 µL of 104.25 TCID₅₀ /100 µL. The morbidity (weight body loss) (a) and mortality (percent survival) (b) rates were monitored within 14 days. (c) virus shedding in mice lungs. Mice were sacrificed three days p.i and lungs were preserved in DMEM and viral loads were detected using TCID₅₀ assay.

Figure 6

ISM analysis of avian HPAIV H5N1 viruses representing different clades emerged in Egypt since 2016. The data confirm the general higher tendency of the 2014 to 2015 HPAIV H5N1 isolated from Egypt towards mammalian-type receptors.