Complex N-glycans are important for interspecies transmission of H7 influenza A viruses

Abstract

Influenza A viruses (IAVs) can overcome species barriers by adaptation of the receptor-binding site of the hemagglutinin (HA). To initiate infection, HAs bind to glycan receptors with terminal sialic acids, which are either N-acetylneuraminic acid (NeuAc) or N-glycolylneuraminic acid (NeuGc); the latter is mainly found in horses and pigs but not in birds and humans. We investigated the influence of previously identified equine NeuGc-adapting mutations (S128T, I130V, A135E, T189A, and K193R) in avian H7 IAVs in vitro and in vivo. We observed that these mutations negatively affected viral replication in chicken cells but not in duck cells and positively affected replication in horse cells. In vivo, the mutations reduced virus virulence and mortality in chickens. Ducks excreted high viral loads longer than chickens, although they appeared clinically healthy. To elucidate why these viruses infected chickens and ducks despite the absence of NeuGc, we re-evaluated the receptor binding of H7 HAs using glycan microarray and flow cytometry studies. This re-evaluation demonstrated that mutated avian H7 HAs also bound to α2,3-linked NeuAc and sialyl-LewisX, which have an additional fucose moiety in their terminal epitope, explaining why infection of ducks and chickens was possible. Interestingly, the α2,3-linked NeuAc and sialyl-LewisX epitopes were only bound when presented on tri-antennary N-glycans, emphasizing the importance of investigating the fine receptor specificities of IAVs. In conclusion, the binding of NeuGc-adapted H7 IAV to tri-antennary N-glycans enables viral replication and shedding by chickens and ducks, potentially facilitating interspecies transmission of equine-adapted H7 IAVs.IMPORTANCEInfluenza A viruses (IAVs) cause millions of deaths and illnesses in birds and mammals each year. The viral surface protein hemagglutinin initiates infection by binding to host cell terminal sialic acids. Hemagglutinin adaptations affect the binding affinity to these sialic acids and the potential host species targeted. While avian and human IAVs tend to bind to N-acetylneuraminic acid (sialic acid), equine H7 viruses prefer binding to N-glycolylneuraminic acid (NeuGc). To better understand the function of NeuGc-specific adaptations in hemagglutinin and to elucidate interspecies transmission potential NeuGc-adapted viruses, we evaluated the effects of NeuGc-specific mutations in avian H7 viruses in chickens and ducks, important economic hosts and reservoir birds, respectively. We also examined the impact on viral replication and found a binding affinity to tri-antennary N-glycans containing different terminal epitopes. These findings are significant as they contribute to the understanding of the role of receptor binding in avian influenza infection.

Keywords: NeuGc; hemagglutinin; influenza A virus; interspecies transmission; sialyl-LewisX.

Fig 1

In vitro characterization of WT (H7N7_avHA) and mutant (H7N7_5eqHA) A/chicken/Germany/R28/2003 H7N7 viruses. (A) Cell-to-cell spread was investigated by measuring the diameter of about 100 plaques in MDCKII cells. (B) Viral replication at indicated time points was assessed in MDCKII, (C) horse lung and horse epidermal cells, (D) chicken fibroblasts (DF-1), primary chicken cells (CEK), SPF embryonated chicken eggs (ECE), and (E) in duck embryo fibroblast cells. (F) The receptor-binding affinity to α2,3-linked NeuAc was measured using α2,3-Sia fetuin substrate. Human H3N2 virus (specific for α2,6-linked NeuAc) was used as a negative control (NC). An avian H4N2 virus (specific for α2,3-linked NeuAc) was used as a positive control (PC). Shown are representative results calculated as means and standard deviations of three independent experiments; each was run in duplicates. (G) pH-dependent activation of HA in a fusion assay was measured after transfection of quail cells (QM-9) with pCAGGS protein expression vector containing the HA of H7N7_avHA or H7N7_5eqHA. Cells were simultaneously transfected with pCAGGS carrying eGFP to facilitate the evaluation of the assay. Cell fusion was triggered 24 hpi with PBS of different pH values for 2 min. The diameter of syncytia was measured using Eclipse Ti-S with the NIS-Elements software version 4.0; Nikon. (H) The thermostability of viruses was measured in duplicates and repeated twice. The reduction in virus infectivity at indicated time points was assessed by titration of heated viruses using a plaque test in MDCKII cells and expressed as plaque-forming units per milliliter (Log10 PFU/mL). All results are expressed as means and standard deviations of at least two independent experiments run in duplicates. t-Test was used to analyze the data in panels C and E, and ANOVA was used for the other panels. Asterisks indicate statistical significance based on P values: *  ≤0.05, **  ≤0.01, *** ≤0.001, **** ≤0.0001, ns, no significance. Dashed lines indicate the predicted detection limit of the plaque assay (cutoff).

Fig 2

In vivo characterization of WT (H7N7_avHA) and mutant (H7N7_5eqHA) A/chicken/Germany/R28/2003 H7N7 virus. (A) Survival and clinical score of ON-infected chickens throughout the animal experiment. (B) Survival curve of ON-infected Pekin ducks. (C) Analysis of oral and cloacal swab samples taken from chickens and ducks in plaque tests expressed as Log10 PFU/mL. (D) The viral distribution in duck and chicken organs was analyzed in plaque tests and expressed as PFU/g. Statistical significance was tested using ANOVA. Asterisks indicate statistical significance based on P values *  ≤0.05, **  ≤0.01, *** ≤0.001, **** ≤0.0001. Dashed lines indicate the predicted detection limit of the plaque assay (cutoff).

Fig 3

Avian H7 HAs bind to sialyl-LewisX epitopes. Synthetic glycans (A) were used to assess the receptor binding of (B) IAVs H7N7_avHA and H7N7_5eq, (C) the IAV HAs, A/turkey/Italy/214845/2002 H7, (D) A/Vietnam/1203/2004 H5, (E) A/duck/Australia/341/1983 H15, and A/Taiwan/2/2013 H6. The glycans were terminating in galactose (no SIA, gray), α2,3-linked NeuAc (white), α2,3-linked NeuGc (blue), α2,6-linked NeuAc (red), or sialyl-LewisX (black). Bars with two colors indicate glycans terminating in different epitopes on different arms. Bars represent the mean ± SD (n = 4).

Fig 4

Avian H7 HAs bind both α2,3-linked NeuAc and sialyl-LewisX epitopes. (A) The presence of sialyl-LewisX epitopes on chicken and duck trachea and colon was investigated using anti-sialyl-LewisX antibodies. (B) The binding of anti-sialyl-LewisX antibodies and H7 HAs to chicken tracheal tissue (with and without fucosidase treatment) was assessed. Tissue binding was visualized using AEC staining. (C) Synthetic glycans (with and without fucosidase treatment) were used to assess the binding of the anti-sialyl-LewisX antibody, (D) H5 HAs of A/chicken/Ibaraki/1/2005, (E) and the WT and mutant (S128T, I130V, A135E, T189A, K193R) H7 HA of A/turkey/Italy/214845/02.