Aberrant Cellular Glycosylation May Increase the Ability of Influenza Viruses to Escape Host Immune Responses through Modification of the Viral Glycome.

Individuals with metabolic dysregulation of cellular glycosylation often experience severe influenza disease, with a poor immune response to the virus and low vaccine efficacy. Here, we investigate the consequences of aberrant cellular glycosylation for the glycome and the biology of influenza virus. We transiently induced aberrant N-linked glycosylation in cultured cells with an oligosaccharyltransferase inhibitor, NGI-1. Cells treated with NGI-1 produced morphologically unaltered viable influenza virus with sequence-neutral glycosylation changes (primarily reduced site occupancy) in the hemagglutinin and neuraminidase proteins. Hemagglutinin with reduced glycan occupancy required a higher concentration of surfactant protein D (an important innate immunity respiratory tract collectin) for inhibition compared to that with normal glycan occupancy. Immunization of mice with NGI-1-treated virus significantly reduced antihemagglutinin and antineuraminidase titers of total serum antibody and reduced hemagglutinin protective antibody responses. Our data suggest that aberrant cellular glycosylation may increase the risk of severe influenza as a result of the increased ability of glycome-modified influenza viruses to evade the immune response. IMPORTANCE People with disorders such as cancer, autoimmune disease, diabetes, or obesity often have metabolic dysregulation of cellular glycosylation and also have more severe influenza disease, a reduced immune response to the virus, and reduced vaccine efficacy. Since influenza viruses that infect such people do not show consistent genomic variations, it is generally assumed that the altered biology is mainly related to host factors. However, since host cells are responsible for glycosylation of influenza virus hemagglutinin and neuraminidase, and glycosylation is important for interactions of these proteins with the immune system, the viruses may have functional differences that are not reflected by their genomic sequence. Here, we show that imbalanced cellular glycosylation can modify the viral glycome without genomic changes, leading to reduced innate and adaptive host immune responses to infection. Our findings link metabolic dysregulation of host glycosylation to increased risk of severe influenza and reduced influenza virus vaccine efficacy.

Influenza virus N-linked glycosylation and innate immunity.

Influenza viruses cause seasonal epidemics and sporadic pandemics in humans. The virus's ability to change its antigenic nature through mutation and recombination, and the difficulty in developing highly effective universal vaccines against it, make it a serious global public health challenge. Influenza virus's surface glycoproteins, hemagglutinin and neuraminidase, are all modified by the host cell's N-linked glycosylation pathways. Host innate immune responses are the first line of defense against infection, and glycosylation of these major antigens plays an important role in the generation of host innate responses toward the virus. Here, we review the principal findings in the analytical techniques used to study influenza N-linked glycosylation, the evolutionary dynamics of N-linked glycosylation in seasonal versus pandemic and zoonotic strains, its role in host innate immune responses, and the prospects for lectin-based therapies. As the efficiency of innate immune responses is a critical determinant of disease severity and adaptive immunity, the study of influenza glycobiology is of clinical as well as research interest.

Virulent PB1-F2 residues: effects on fitness of H1N1 influenza A virus in mice and changes during evolution of human influenza A viruses.

Specific residues of influenza A virus (IAV) PB1-F2 proteins may enhance inflammation or cytotoxicity. In a series of studies, we evaluated the function of these virulence-associated residues in the context of different IAV subtypes in mice. Here, we demonstrate that, as with the previously assessed pandemic 1968 (H3N2) IAV, PB1-F2 inflammatory residues increase the virulence of H1N1 IAV, suggesting that this effect might be a universal feature. Combining both inflammatory and cytotoxic residues in PB1-F2 enhanced virulence further, compared to either motif alone. Residues from these virulent motifs have been present in natural isolates from human seasonal IAV of all subtypes, but there has been a trend toward a gradual reduction in the number of virulent residues over time. However, human IAV of swine and avian origin tend to have more virulent residues than do the human-adapted seasonal strains, raising the possibility that donation of PB1 segments from these zoonotic viruses may increase the severity of some seasonal human strains. Our data suggest the value of surveillance of virulent residues in both human and animal IAV to predict the severity of influenza season.

Evolution and Virulence of Influenza A Virus Protein PB1-F2.

PB1-F2 is an accessory protein of most human, avian, swine, equine, and canine influenza A viruses (IAVs). Although it is dispensable for virus replication and growth, it plays significant roles in pathogenesis by interfering with the host innate immune response, inducing death in immune and epithelial cells, altering inflammatory responses, and promoting secondary bacterial pneumonia. The effects of PB1-F2 differ between virus strains and host species. This can at least partially be explained by the presence of multiple PB1-F2 sequence variants, including premature stop codons that lead to the expression of truncated PB1-F2 proteins of different lengths and specific virulence-associated residues that enhance susceptibility to bacterial superinfection. Although there has been a tendency for human seasonal IAV to gradually reduce the number of virulence-associated residues, zoonotic IAVs contain a reservoir of PB1-F2 proteins with full length, virulence-associated sequences. Here, we review the molecular mechanisms by which PB1-F2 may affect influenza virulence, and factors associated with the evolution and selection of this protein.

Glycosylation changes in the globular head of H3N2 influenza hemagglutinin modulate receptor binding without affecting virus virulence.

Since the emergence of human H3N2 influenza A viruses in the pandemic of 1968, these viruses have become established as strains of moderate severity. A decline in virulence has been accompanied by glycan accumulation on the hemagglutinin globular head, and hemagglutinin receptor binding has changed from recognition of a broad spectrum of glycan receptors to a narrower spectrum. The relationship between increased glycosylation, binding changes, and reduction in H3N2 virulence is not clear. We evaluated the effect of hemagglutinin glycosylation on receptor binding and virulence of engineered H3N2 viruses. We demonstrate that low-binding virus is as virulent as higher binding counterparts, suggesting that H3N2 infection does not require either recognition of a wide variety of, or high avidity binding to, receptors. Among the few glycans recognized with low-binding virus, there were two structures that were bound by the vast majority of H3N2 viruses isolated between 1968 and 2012. We suggest that these two structures support physiologically relevant binding of H3N2 hemagglutinin and that this physiologically relevant binding has not changed since the 1968 pandemic. Therefore binding changes did not contribute to reduced severity of seasonal H3N2 viruses. This work will help direct the search for factors enhancing influenza virulence.

Non-avian animal reservoirs present a source of influenza A PB1-F2 proteins with novel virulence-enhancing markers.

PB1-F2 protein, expressed from an alternative reading frame of most influenza A virus (IAV) PB1 segments, may possess specific residues associated with enhanced inflammation (L62, R75, R79, and L82) and cytotoxicity (I68, L69, and V70). These residues were shown to increase the pathogenicity of primary viral and secondary bacterial infections in a mouse model. In contrast to human seasonal influenza strains, virulence-associated residues are present in PB1-F2 proteins from pandemic H1N1 1918, H2N2 1957, and H3N2 1968, and highly pathogenic H5N1 strains, suggesting their contribution to viruses' pathogenic phenotypes. Non-human influenza strains may act as donors of virulent PB1-F2 proteins. Previously, avian influenza strains were identified as a potential source of inflammatory, but not cytotoxic, PB1-F2 residues. Here, we analyze the frequency of virulence-associated residues in PB1-F2 sequences from IAVs circulating in mammalian species in close contact with humans: pigs, horses, and dogs. All four inflammatory residues were found in PB1-F2 proteins from these viruses. Among cytotoxic residues, I68 was the most common and was especially prevalent in equine and canine IAVs. Historically, PB1-F2 from equine (about 75%) and canine (about 20%) IAVs were most likely to have combinations of the highest numbers of residues associated with inflammation and cytotoxicity, compared to about 7% of swine IAVs. Our analyses show that, in addition to birds, pigs, horses, and dogs are potentially important sources of pathogenic PB1-F2 variants. There is a need for surveillance of IAVs with genetic markers of virulence that may be emerging from these reservoirs in order to improve pandemic preparedness and response.

A novel cytotoxic sequence contributes to influenza A viral protein PB1-F2 pathogenicity and predisposition to secondary bacterial infection.

Enhancement of cell death is a distinguishing feature of H1N1 influenza virus A/Puerto Rico/8/34 protein PB1-F2. Comparing the sequences (amino acids [aa] 61 to 87 using PB1-F2 amino acid numbering) of the PB1-F2-derived C-terminal peptides from influenza A viruses inducing high or low levels of cell death, we identified a unique I68, L69, and V70 motif in A/Puerto Rico/8/34 PB1-F2 responsible for promotion of the peptide's cytotoxicity and permeabilization of the mitochondrial membrane. When administered to mice, a 27-mer PB1-F2-derived C-terminal peptide with this amino acid motif caused significantly greater weight loss and pulmonary inflammation than the peptide without it (due to I68T, L69Q, and V70G mutations). Similar to the wild-type peptide, A/Puerto Rico/8/34 elicited significantly higher levels of macrophages, neutrophils, and cytokines in the bronchoalveolar lavage fluid of mice than its mutant counterpart 7 days after infection. Additionally, infection of mice with A/Puerto Rico/8/34 significantly enhanced the levels of morphologically transformed epithelial and immune mononuclear cells recruited in the airways compared with the mutant virus. In the mouse bacterial superinfection model, both peptide and virus with the I68, L69, and V70 sequence accelerated development of pneumococcal pneumonia, as reflected by increased levels of viral and bacterial lung titers and by greater mortality. Here we provide evidence suggesting that the newly identified cytotoxic sequence I68, L69, and V70 of A/Puerto Rico/8/34 PB1-F2 contributes to the pathogenesis of both primary viral and secondary bacterial infections.

Receptor-binding specificity of the human parainfluenza virus type 1 hemagglutinin-neuraminidase glycoprotein.

The hemagglutinin-neuraminidase (HN) glycoprotein is utilized by human parainfluenza viruses for binding to the host cell. By the use of glycan array assays, we demonstrate that, in addition to the first catalytic-binding site, the HN of human parainfluenza virus type 1 has a second site for binding covered by N-linked glycan. Our data suggest that attachment of the first site to sialic acid (SA)-linked receptors triggers exposure of the second site. We found that both sites bind to α2-3-linked SAs with a preference for a sialyl-Lewis(x) motif. Binding to α2-3-linked SAs with a sulfated sialyl-Lewis motif as well as to α2-8-linked SAs was unique for the second binding site. Neither site recognizes α2-6-linked oligosaccharides.

Immunopathogenic and antibacterial effects of H3N2 influenza A virus PB1-F2 map to amino acid residues 62, 75, 79, and 82.

The influenza A virus protein PB1-F2 has been linked to the pathogenesis of both primary viral and secondary bacterial infections. H3N2 viruses have historically expressed full-length PB1-F2 proteins with either proinflammatory (e.g., from influenza A/Hong Kong/1/1968 virus) or noninflammatory (e.g., from influenza A/Wuhan/359/1995 virus) properties. Using synthetic peptides derived from the active C-terminal portion of the PB1-F2 protein from those two viruses, we mapped the proinflammatory domain to amino acid residues L62, R75, R79, and L82 and then determined the role of that domain in H3N2 influenza virus pathogenicity. PB1-F2-derived peptides containing that proinflammatory motif caused significant morbidity, mortality, and pulmonary inflammation in mice, manifesting as increased acute lung injury and the presence of proinflammatory cytokines and inflammatory cells in the lungs compared to peptides lacking this motif, and better supported bacterial infection with Streptococcus pneumoniae. Infections of mice with an otherwise isogenic virus engineered to contain this proinflammatory sequence in PB1-F2 demonstrated increased morbidity resulting from primary viral infections and enhanced development of secondary bacterial pneumonia. The presence of the PB1-F2 noninflammatory (P62, H75, Q79, and S82) sequence in the wild-type virus mediated an antibacterial effect. These data suggest that loss of the inflammatory PB1-F2 phenotype that supports bacterial superinfection during adaptation of H3N2 viruses to humans, coupled with acquisition of antibacterial activity, contributes to the relatively diminished frequency of severe infections seen with seasonal H3N2 influenza viruses in recent decades compared to their first 2 decades of circulation.

N-linked glycan at residue 523 of human parainfluenza virus type 3 hemagglutinin-neuraminidase masks a second receptor-binding site.

The hemagglutinin-neuraminidase (HN) glycoprotein plays a critical role in parainfluenza virus replication. We recently found that in addition to the catalytic binding site, HN of human parainfluenza virus type 1 (hPIV-1) may have a second receptor-binding site covered by an N-linked glycan at residue 173, which is near the region of the second receptor-binding site identified in Newcastle disease virus (NDV) HN (I. A. Alymova, G. Taylor, V. P. Mishin, M. Watanabe, K. G. Murti, K. Boyd, P. Chand, Y. S. Babu, and A. Portner, J. Virol. 82:8400-8410, 2008). Sequence analysis and superposition of the NDV and hPIV-3 HN dimer structures revealed that, similar to what was seen in hPIV-1, the N-linked glycan at residue 523 on hPIV-3 HN may cover a second receptor-binding site. Removal of this N-linked glycosylation site by an Asn-to-Asp substitution at residue 523 (N523D) changed the spectrum of the mutant virus's receptor specificity, delayed its elution from both turkey and chicken red blood cells, reduced mutant sensitivity (by about half) to the selective HN inhibitor BCX 2855 in hemagglutination inhibition tests, and slowed its growth in LLC-MK(2) cells. The neuraminidase activity of the mutant and its sensitivity to BCX 2855 in neuraminidase inhibition assays did not change, indicating that the mutation did not affect the virus's catalytic-binding site and that all observed effects were caused by the exposure of the purported second receptor-binding site. Our data are consistent with the idea that, similar to the case for hPIV-1, the N-linked glycan shields a second receptor-binding site on hPIV-3 HN.

Efficacy of the novel parainfluenza virus haemagglutinin-neuraminidase inhibitor BCX 2798 in mice - further evaluation.

BACKGROUND: Human parainfluenza virus type 1 (hPIV-1) causes serious respiratory tract infections, especially in children. This study investigated the efficacy of the novel haemagglutinin-neuraminidase (HN) inhibitor BCX 2798 in the prophylaxis of lethal and the treatment of non-lethal parainfluenza virus infection in mice. METHODS: In the prophylaxis model, 129x1/SvJ mice were inoculated with a 90% lethal dose of a recombinant Sendai virus, in which the HN gene was replaced with that of hPIV-1 (rSeV[hPIV-1HN]). The mice were intranasally treated either once or for 5 days with 1 or 10 mg/kg/day of BCX 2798, starting 4 h before infection. In the therapeutic model, mice were infected with 100 plaque-forming units of rSeV(hPIV-1HN) per mouse and treated intranasally with 0.1, 1 or 10 mg/kg/day of BCX 2798 for 5 days, starting 24 or 48 h after infection, or for 4 days starting 72 h after infection. RESULTS: Similar to multiple dosing, a single intranasal prophylaxis with 1 or 10 mg/kg of BCX 2798 protected approximately 40% or 90%, respectively, of mice from death by rSeV(hPIV-1HN) infection. BCX 2798 also significantly reduced virus lung titres (in a dose- and time-dependent manner) and reduced histopathological changes in the airways of non-lethally infected mice at multiple intranasal dosages in the therapeutic model, with the lowest effective dosage being 0.1 mg/kg/day administered 24 h after infection. CONCLUSIONS: BCX 2798 was effective in the prophylaxis of lethal and in the therapy of non-lethal parainfluenza virus infection in mice, suggesting further consideration of BCX 2798 for clinical trials.

Effect of hemagglutinin-neuraminidase inhibitors BCX 2798 and BCX 2855 on growth and pathogenicity of Sendai/human parainfluenza type 3 chimera virus in mice.

Human parainfluenza virus type 3 (hPIV-3) is a major respiratory tract pathogen that affects young children, but no vaccines or antiviral drugs against it have yet been developed. We developed a mouse model to evaluate the efficacies of the novel parainfluenza virus hemagglutinin-neuraminidase (HN) inhibitors BCX 2798 and BCX 2855 against a recombinant Sendai virus (rSeV) in which the fusion (F) and HN surface glycoproteins (FHN) were replaced by those of hPIV-3 [rSeV(hPIV-3FHN)]. In the prophylaxis model, 129X1/SvJ mice were infected with a 90% or 20% lethal dose of the virus and were treated intranasally for 5 days with 10 mg/kg of body weight/day of either compound starting 4 h before infection. Prophylactic treatment of the mice with either compound did not prevent their death in a 90% lethality model of rSeV(hPIV-3FHN) infection. However, it significantly reduced the lung virus titers, the amount of weight lost, and the rate of mortality in mice infected with a 20% lethal virus dose. In the therapy model, mice were infected with a nonlethal dose of the virus (100 PFU/mouse) and were treated intranasally with 1 or 10 mg/kg/day of either compound for 5 days starting at 24 or 48 h postinfection. Treatment of the mice with either compound significantly reduced the virus titer in the lungs, subsequently causing a reduction in the number of immune cells and the levels of cytokines in the bronchoalveolar lavage fluid and histopathologic changes in the airways. Our results indicate that BCX 2798 and BCX 2855 are effective inhibitors of hPIV-3 HN in our mouse model and may be promising candidates for the prophylaxis and treatment of hPIV-3 infection in humans.

Loss of the N-linked glycan at residue 173 of human parainfluenza virus type 1 hemagglutinin-neuraminidase exposes a second receptor-binding site.

BCX 2798 (4-azido-5-isobutyrylamino-2,3-didehydro-2,3,4,5-tetradeoxy-d-glycero-d-galacto-2-nonulopyranosic acid) effectively inhibited the activities of the hemagglutinin-neuraminidase (HN) of human parainfluenza viruses (hPIV) in vitro and protected mice from lethal infection with a recombinant Sendai virus whose HN was replaced with that of hPIV-1 (rSeV[hPIV-1HN]) (I. V. Alymova, G. Taylor, T. Takimoto, T. H. Lin., P. Chand, Y. S. Babu, C. Li, X. Xiong, and A. Portner, Antimicrob. Agents Chemother. 48:1495-1502, 2004). The ability of BCX 2798 to select drug-resistant variants in vivo was examined. A variant with an Asn-to-Ser mutation at residue 173 (N173S) in HN was recovered from mice after a second passage of rSeV(hPIV-1HN) in the presence of BCX 2798 (10 mg/kg of body weight daily). The N173S mutant remained sensitive to BCX 2798 in neuraminidase inhibition assays but was more than 10,000-fold less sensitive to the compound in hemagglutination inhibition tests than rSeV(hPIV-1HN). Its susceptibility to BCX 2798 in plaque reduction assays was reduced fivefold and did not differ from that of rSeV(hPIV-1HN) in mice. The N173S mutant failed to be efficiently eluted from erythrocytes and released from cells. It demonstrated reduced growth in cell culture and superior growth in mice. The results for gel electrophoresis analysis were consistent with the loss of the N-linked glycan at residue 173 in the mutant. Sequence and structural comparisons revealed that residue 173 on hPIV-1 HN is located close to the region of the second receptor-binding site identified in Newcastle disease virus HN. Our study suggests that the N-linked glycan at residue 173 masks a second receptor-binding site on hPIV-1 HN.

The novel parainfluenza virus hemagglutinin-neuraminidase inhibitor BCX 2798 prevents lethal synergism between a paramyxovirus and Streptococcus pneumoniae.

An association exists between respiratory viruses and bacterial infections. Prevention or treatment of the preceding viral infection is a logical goal for reducing this important cause of morbidity and mortality. The ability of the novel, selective parainfluenza virus hemagglutinin-neuraminidase inhibitor BCX 2798 to prevent the synergism between a paramyxovirus and Streptococcus pneumoniae was examined in this study. A model of secondary bacterial pneumonia after infection with a recombinant Sendai virus whose hemagglutinin-neuraminidase gene was replaced with that of human parainfluenza virus type 1 [rSV(hHN)] was established in mice. Challenge of mice with a sublethal dose of S. pneumoniae 7 days after a sublethal infection with rSV(hHN) (synergistic group) caused 100% mortality. Bacterial infection preceding viral infection had no effect on survival. The mean bacterial titers in the synergistic group were significantly higher than in mice infected with bacteria only. The virus titers were similar in mice infected with rSV(hHN) alone and in dually infected mice. Intranasal administration of BCX 2798 at 10 mg/kg per day to the synergistic group of mice starting 4 h before virus infection protected 80% of animals from death. This effect was accompanied by a significant reduction in lung viral and bacterial titers. Treatment of mice 24 h after the rSV(hHN) infection showed no protection against synergistic lethality. Together, our results indicate that parainfluenza viruses can prime for secondary bacterial infections. Prophylaxis of parainfluenza virus infections with antivirals might be an effective strategy for prevention of secondary bacterial complications in humans.

Efficacy of novel hemagglutinin-neuraminidase inhibitors BCX 2798 and BCX 2855 against human parainfluenza viruses in vitro and in vivo.

Human parainfluenza viruses are important respiratory tract pathogens, especially of children. However, no vaccines or specific therapies for infections caused by these viruses are currently available. In the present study we characterized the efficacy of the novel parainfluenza virus inhibitors BCX 2798 and BCX 2855, which were designed based on the three-dimensional structure of the hemagglutinin-neuraminidase (HN) protein. The compounds were highly effective in inhibiting hemagglutinin (HA) and neuraminidase (NA) activities and the growth of hPIV-1, hPIV-2, and hPIV-3 in LLC-MK(2) cells. The concentrations required to reduce the activity to 50% of that of a control ranged from 0.1 to 6.0 micro M in HA inhibition assays and from 0.02 to 20 micro M in NA inhibition assays. The concentrations required to inhibit virus replication to 50% of the level of the control ranged from 0.7 to 11.5 micro M. BCX 2798 and BCX 2855 were inactive against influenza virus HA and NA and bacterial NA. In mice infected with a recombinant Sendai virus whose HN gene was replaced with that of hPIV-1 [rSV(hHN)], intranasal administration of BCX 2798 (10 mg/kg per day) and of BCX 2855 (50 mg/kg per day) 4 h before the start of infection resulted in a significant reduction in titers of virus in the lungs and protection from death. Treatment beginning 24 h after the start of infection did not prevent death. Together, our results indicate that BCX 2798 and BCX 2855 are effective inhibitors of parainfluenza virus HN and may limit parainfluenza virus infections in humans.