Positive selection for gains of N-linked glycosylation sites in hemagglutinin during evolution of H3N2 human influenza A virus

Analysis of mutation-originated gain-of-glycosylation using mass spectrometry-based N-glycoproteomics

RATIONALE

A general N-glycoproteomics analysis pipeline has been established for characterization of mutation-related gain-of-glycosylation (GoG) at intact N-glycopeptide molecular level, generating comprehensive site and structure information of N-glycosylation.

METHODS

This study focused on mutation-originated GoG using a mass spectrometry-based N-glycoproteomics analysis workflow. In brief, GoG intact N-glycopeptide databases were built, consisting of 2701 proteins (potential GoG N-glycosites and amino acids derived from MUTAGEN, VARIANT and VAR_SEQ in UniProt) and 6709 human N-glycans (≤50 sequence isomers per monosaccharide composition). We employed the site- and structure-specific N-glycoproteomics workflow utilizing intact N-glycopeptides search engine GPSeeker to identify GoG intact N-glycopeptides from parental breast cancer stem cells (MCF-7 CSCs) and adriamycin-resistant breast cancer stem cells (MCF-7/ADR CSCs).

RESULTS

With the criteria of spectrum-level false discovery rate control of ≤1%, we identified 87 and 94 GoG intact N-glycopeptides corresponding to 37 and 35 intact N-glycoproteins from MCF-7 CSCs and MCF-7/ADR CSCs, respectively. Micro-heterogeneity and macro-heterogeneity of N-glycosylation from GoG intact N-glycoproteins with VAR_SEQ and VARIANT were found in both MCF-7 CSCs and MCF-7/ADR CSCs systems.

CONCLUSIONS

The integration of site- and structure-specific N-glycoproteomics approach, conjugating with GoG characterization, provides a universal workflow for revealing comprehensive N-glycosite and N-glycan structure information of GoG. The analysis of mutation-originated GoG can be extended to GoG characterization of other N-glycoproteome systems including complex clinical tissues and body fluids.

Combining machine learning with structure-based protein design to predict and engineer post-translational modifications of proteins

Post-translational modifications (PTMs) of proteins play a vital role in their function and stability. These modifications influence protein folding, signaling, protein-protein interactions, enzyme activity, binding affinity, aggregation, degradation, and much more. To date, over 400 types of PTMs have been described, representing chemical diversity well beyond the genetically encoded amino acids. Such modifications pose a challenge to the successful design of proteins, but also represent a major opportunity to diversify the protein engineering toolbox. To this end, we first trained artificial neural networks (ANNs) to predict eighteen of the most abundant PTMs, including protein glycosylation, phosphorylation, methylation, and deamidation. In a second step, these models were implemented inside the computational protein modeling suite Rosetta, which allows flexible combination with existing protocols to model the modified sites and understand their impact on protein stability as well as function. Lastly, we developed a new design protocol that either maximizes or minimizes the predicted probability of a particular site being modified. We find that this combination of ANN prediction and structure-based design can enable the modification of existing, as well as the introduction of novel, PTMs. The potential applications of our work include, but are not limited to, glycan masking of epitopes, strengthening protein-protein interactions through phosphorylation, as well as protecting proteins from deamidation liabilities. These applications are especially important for the design of new protein therapeutics where PTMs can drastically change the therapeutic properties of a protein. Our work adds novel tools to Rosetta's protein engineering toolbox that allow for the rational design of PTMs.

Chimeric hemagglutinin split vaccines elicit broadly cross-reactive antibodies and protection against group 2 influenza viruses in mice

Seasonal influenza virus vaccines are effective when they are well matched to circulating strains. Because of antigenic drift/change in the immunodominant hemagglutinin (HA) head domain, annual vaccine reformulations are necessary to maintain a match with circulating strains. In addition, seasonal vaccines provide little to no protection against newly emerging pandemic strains. Sequential vaccination with chimeric HA (cHA) constructs has been proven to direct the immune response toward the immunosubdominant but more conserved HA stalk domain. In this study, we show that immunization with group 2 cHA split vaccines in combination with the CpG 1018 adjuvant elicits broadly cross-reactive antibodies against all group 2 HAs, as well as systemic and local antigen-specific T cell responses. Antibodies elicited after sequential vaccination are directed to conserved regions of the HA such as the stalk and the trimer interface and also to the N2 neuraminidase (NA). Immunized mice were fully protected from challenge with a broad panel of influenza A viruses.

Influenza Vaccine Effectiveness: Defining the H3N2 Problem

Observational studies have consistently shown that influenza vaccine effectiveness (VE) is lower for H3N2 relative to H1N1pdm09 and type B, and this is not entirely explained by antigenic match. The triad of virus, vaccine, and host immunity provides a framework to examine contributing factors. Antigenic evolution facilitates H3N2 immune escape, and increasing glycosylation of the hemagglutinin shields antigenic sites from antibody binding. Egg passage adaptation of vaccine viruses generates mutations that alter glycosylation, impair the neutralizing antibody response, and reduce VE. Complex host immune factors may also influence H3N2 VE, including early childhood imprinting and repeated vaccination, but their role is uncertain. Of the triad of contributing factors, only changes to the vaccine are readily achievable. However, it is unclear whether current licensed non–egg-based vaccines generate superior protection against H3N2. The optimal strategy remains to be defined, but newer vaccine technology platforms offer great potential.

Efficacy of novel recombinant fowlpox vaccine against recent Mexican H7N3 highly pathogenic avian influenza virus

Since 2012, H7N3 highly pathogenic avian influenza (HPAI) has produced negative economic and animal welfare impacts on poultry in central Mexico. In the present study, chickens were vaccinated with two different recombinant fowlpox virus vaccines (rFPV-H7/3002 with 2015 H7 hemagglutinin [HA] gene insert, and rFPV-H7/2155 with 2002 H7 HA gene insert), and were then challenged three weeks later with H7N3 HPAI virus (A/chicken/Jalisco/CPA-37905/2015). The rFPV-H7/3002 vaccine conferred 100% protection against mortality and morbidity, and significantly reduced virus shed titers from the respiratory and gastrointestinal tracts. In contrast, 100% of sham and rFPV-H7/2155 vaccinated birds shed virus at higher titers and died within 4 days. Pre- (15/20) and post- (20/20) challenge serum of birds vaccinated with rFPV-H7/3002 had antibodies detectable by hemagglutination inhibition (HI) assay using challenge virus antigen. However, only a few birds (3/20) in the rFPV-H7/2155 vaccinated group had antibodies that reacted against the challenge strain but all birds had antibodies that reacted against the homologous vaccine antigen (A/turkey/Virginia/SEP-66/2002) (20/20). One possible explanation for differences in vaccines efficacy is the antigenic drift between circulating viruses and vaccines. Molecular analysis demonstrated that the Mexican H7N3 strains have continued to rapidly evolve since 2012. In addition, we identified in silico three potential new N-glycosylation sites on the globular head of the H7 HA of A/chicken/Jalisco/CPA-37905/2015 challenge virus, which were absent in 2012 H7N3 outbreak virus. Our results suggested that mutations in the HA antigenic sites including increased glycosylation sites, accumulated in the new circulating Mexican H7 HPAIV strains, altered the recognition of neutralizing antibodies from the older vaccine strain rFPV-H7/2155. Therefore, the protective efficacy of novel rFPV-H7/3002 against recent outbreak Mexican H7N3 HPAIV confirms the importance of frequent updating of vaccines seed strains for long-term effective control of H7 HPAI virus.

Influenza A surface glycosylation and vaccine design

We have shown that glycosylation of influenza A virus (IAV) hemagglutinin (HA), especially at position N-27, is crucial for HA folding and virus survival. However, it is not known whether the glycosylation of HA and the other two major IAV surface glycoproteins, neuraminidase (NA) and M2 ion channel, is essential for the replication of IAV. Here, we show that glycosylation of HA at N-142 modulates virus infectivity and host immune response. Glycosylation of NA in the stalk region affects its structure, activity, and specificity, thereby modulating virus release and virulence, and glycosylation at the catalytic domain affects its thermostability; however, glycosylation of M2 had no effect on its function. In addition, using IAV without the stalk and catalytic domains of NA as a live attenuated vaccine was shown to confer a strong IAV-specific CD8⁺ T-cell response and a strong cross-strain as well as cross-subtype protection against various virus strains.

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.

Assessing Antigenic Drift of Seasonal Influenza A(H3N2) and A(H1N1)pdm09 Viruses

Under selective pressure from the host immune system, antigenic epitopes of influenza virus hemagglutinin (HA) have continually evolved to escape antibody recognition, termed antigenic drift. We analyzed the genomes of influenza A(H3N2) and A(H1N1)pdm09 virus strains circulating in Thailand between 2010 and 2014 and assessed how well the yearly vaccine strains recommended for the southern hemisphere matched them. We amplified and sequenced the HA gene of 120 A(H3N2) and 81 A(H1N1)pdm09 influenza virus samples obtained from respiratory specimens and calculated the perfect-match vaccine efficacy using the p_epitope model, which quantitated the antigenic drift in the dominant epitope of HA. Phylogenetic analysis of the A(H3N2) HA1 genes classified most strains into genetic clades 1, 3A, 3B, and 3C. The A(H3N2) strains from the 2013 and 2014 seasons showed very low to moderate vaccine efficacy and demonstrated antigenic drift from epitopes C and A to epitope B. Meanwhile, most A(H1N1)pdm09 strains from the 2012–2014 seasons belonged to genetic clades 6A, 6B, and 6C and displayed the dominant epitope mutations at epitopes B and E. Finally, the vaccine efficacy for A(H1N1)pdm09 (79.6–93.4%) was generally higher than that of A(H3N2). These findings further confirmed the accelerating antigenic drift of the circulating influenza A(H3N2) in recent years.

K-Pax2: Bayesian identification of cluster-defining amino acid positions in large sequence datasets

The recent growth in publicly available sequence data has introduced new opportunities for studying microbial evolution and spread. Because the pace of sequence accumulation tends to exceed the pace of experimental studies of protein function and the roles of individual amino acids, statistical tools to identify meaningful patterns in protein diversity are essential. Large sequence alignments from fast-evolving micro-organisms are particularly challenging to dissect using standard tools from phylogenetics and multivariate statistics because biologically relevant functional signals are easily masked by neutral variation and noise. To meet this need, a novel computational method is introduced that is easily executed in parallel using a cluster environment and can handle thousands of sequences with minimal subjective input from the user. The usefulness of this kind of machine learning is demonstrated by applying it to nearly 5000 haemagglutinin sequences of influenza A/H3N2.Antigenic and 3D structural mapping of the results show that the method can recover the major jumps in antigenic phenotype that occurred between 1968 and 2013 and identify specific amino acids associated with these changes. The method is expected to provide a useful tool to uncover patterns of protein evolution.

Predictability of antigenic evolution for H3N2 human influenza A virus

Influenza A virus continues to pose a threat to public health. Since this virus can evolve escape mutants rapidly, it is desirable to predict the antigenic evolution for developing effective vaccines. Although empirical methods have been proposed and reported to predict the antigenic evolution more or less accurately, they did not provide much insight into the effects of unobserved mutations and the mechanisms of antigenic evolution. Here a theoretical method was introduced to predict the antigenic evolution of H3N2 human influenza A virus by evaluating de novo mutations through estimating the antigenic distance. The antigenic distance defined with the hemagglutination inhibition (HI) titer was estimated with antigenic models taking into account the volume, isoelectric point, relative solvent accessibility, and distances from receptor-binding sites (RBS) and N-linked glycosylation sites (NGS) for amino acids in hemagglutinin 1 (HA1). When the best model with the optimized parameter values was used to predict the antigenic evolution for the dominant strains, the prediction accuracy was relatively low. However, there appeared to be an overall tendency that the amino acid sites with larger potential net effect on antigenicity were more likely to evolve and the amino acid changes with larger potential effect were more likely to take place, suggesting that natural selection may operate to enhance the antigenic evolution of H3N2 human influenza A virus.

Prevalence, genetic drift of haemagglutinin, and antiviral resistance of influenza A/H3N2 viruses circulating in Shanghai in children during 2009-2012

Influenza A/H3N2 viruses are associated with severe epidemics. Antiviral resistance and continued antigenic drift are the major concerns regarding prophylaxis and treatment of influenza. The objectives of this study were to investigate the prevalence and frequency of antiviral drug resistance in influenza A/H3N2 viruses circulating among Shanghainese children from June 2009 to May 2012 and to understand the genetic evolution of the hemagglutinin (HA) epitopes. Nasopharyngeal/throat swabs were collected from outpatients with influenza-like illness. Of the 3,475 children tested, 344 (9.9%) were positive for influenza A/H3N2 viruses. Epidemics of influenza A/H3N2 occurred in July-September 2009, August 2010-January 2011, and November 2011-May 2012. The 71 A/H3N2-positive specimens were sequenced to characterize the genotypic antiviral resistance and genetic drift in the HA epitopes. All of the 71 A/H3N2 viruses sequenced were genotypically resistant to adamantine but sensitive to oseltamivir. The HA1 sequence analysis revealed that the A/H3N2 viruses underwent constant mutations in the HA antigenic sites over the three seasons compared with the corresponding vaccine strains, and amino acid changes in at least three epitopes were observed each season. Phylogenic analyses indicated that the A/H3N2 strains circulating in Shanghai fell into clades different from those of the corresponding seasonal vaccine strains and were grouped into the A/Perth/16/2009 genetic clade and the A/Victoria/208/2009 genetic clades 3B, 3C, and 5. The continuous monitoring of genetic drift and antiviral resistance of influenza viruses is important for the management of influenza and for updating the vaccine composition.

Genetic mutations in influenza H3N2 viruses from a 2012 epidemic in Southern China

Background

An influenza H3N2 epidemic occurred throughout Southern China in 2012.

Methods

We analyzed the hemagglutinin (HA) and neuraminidase (NA) genes of influenza H3N2 strains isolated between 2011–2012 from Guangdong. Mutation sites, evolutionary selection, antigenic sites, and N-glycosylation within these strains were analyzed.

Results

The 2011–2012 Guangdong strains contained the HA-A214S, HA-V239I, HA-N328S, NA-L81P, and NA-D93G mutations, similar to those seen in the A/ Perth/16/2009 influenza strain. The HA-NSS_061–063 and NNS_160–162 glycosylation sites were prevalent among the 2011–2012 Guangdong strains but the NA-NRS_402–404 site was deleted. Antigenically, there was a four-fold difference between A/Perth/16/2009 -like strains and the 2011–2012 Guangdong strains.

Conclusion

Antigenic drift of the H3N2 subtype contributed to the occurrence of the Southern China influenza epidemic of 2012.

Loss and gain of N-linked glycosylation sites in globular head and stem of HA found in A/H3N2 flu fatal and severe cases during 2013 Tunisia flu seasonal survey

Glycosylation on the globular head of the hemagglutinin (HA) protein of influenza virus acts as an important target for recognition and destruction of virus by innate immune proteins of the collectin family. In the current study, we have characterized the dynamic amino acid changes at N-linked glycosylation sites of full length sequences of HA genes of 5 A/H3N2 Tunisian strains isolates from mild, severe, and fatal cases. Compared to the reference strain, A/Perth/16/2009 substitutions in potential N-glycosylation sites were observed in 5 HA genes at five different positions (45, 124, 128, 144, and 145) generating the losses and gains of N-linked glycosylation sites. Also the mutation N145S was presented in the receptor-binding site of all segments analyzed. Point mutations in several positions in the gene encoding the H3 of Tunisian strains were shown to ablate a glycan attachment site and also loss of a potential glycosylation site. The relation between these mutations and virulence of influenza A/H3N2 virus needed to be verified in the further experiments.

Guiding the immune response against influenza virus hemagglutinin toward the conserved stalk domain by hyperglycosylation of the globular head domain

Influenza virus hemagglutinin consists of a highly variable and immunodominant head domain and a more conserved but immunosubdominant stalk domain. We introduced seven N-linked glycosylation sites in the hemagglutinin head domain to shield the immunodominant antigenic sites. The hyperglycosylated hemagglutinin enhanced stalk-directed seroreactivity while dampening the head response in immunized mice. Upon influenza virus challenge, mice vaccinated with the hyperglycosylated hemagglutinin were better protected against morbidity and mortality than mice receiving the wild-type hemagglutinin.

Age-specific mortality during the 1918 influenza pandemic: unravelling the mystery of high young adult mortality

The worldwide spread of a novel influenza A (H1N1) virus in 2009 showed that influenza remains a significant health threat, even for individuals in the prime of life. This paper focuses on the unusually high young adult mortality observed during the Spanish flu pandemic of 1918. Using historical records from Canada and the U.S., we report a peak of mortality at the exact age of 28 during the pandemic and argue that this increased mortality resulted from an early life exposure to influenza during the previous Russian flu pandemic of 1889-90. We posit that in specific instances, development of immunological memory to an influenza virus strain in early life may lead to a dysregulated immune response to antigenically novel strains encountered in later life, thereby increasing the risk of death. Exposure during critical periods of development could also create holes in the T cell repertoire and impair fetal maturation in general, thereby increasing mortality from infectious diseases later in life. Knowledge of the age-pattern of susceptibility to mortality from influenza could improve crisis management during future influenza pandemics.

Detection of positive selection eliminating effects of structural constraints in hemagglutinin of H3N2 human influenza A virus

In the evolutionary studies of proteins, the average effect of natural selection operating on amino acid mutations may be examined by comparing the numbers of synonymous (dS) and nonsynonymous (dN) substitutions that have accumulated during the same time period. In this method, destabilizing mutations occurring across protein molecules may interfere with detection of natural selection, particularly positive selection, operating on other mutations. Here an attempt to detect positive selection eliminating effects of structural constraints is demonstrated using hemagglutinin (HA) of H3N2 human influenza A virus as an example. Compatible and incompatible amino acids were inferred at each site from the computational analysis of three-dimensional structure using the thermodynamic stability as an indicator, and natural selection was examined by comparing dS and dN among compatible amino acids. In the analysis of 2701 nucleotide sequences for the entire coding region of HA, the new method identified twice as many positively selected amino acid sites as the ordinary method (16 and 4 sites in the former method without and with correction for multiple testing, respectively, and 8 and 2 sites in the latter method). Positively selected sites were involved in epitopes, receptor-binding pocket, epistasis, and stabilization, which appeared to be biologically reasonable. Nevertheless, there still appeared to be several problems, which may largely render this method conservative. It may be effective to analyze many densely sampled sequences in this method.

The evolutionary pattern of glycosylation sites in influenza virus (H5N1) hemagglutinin and neuraminidase

Two glycoproteins, hemagglutinin (HA) and neuraminidase (NA), on the surface of influenza viruses play crucial roles in transfaunation, membrane fusion and the release of progeny virions. To explore the distribution of N-glycosylation sites (glycosites) in these two glycoproteins, we collected and aligned the amino acid sequences of all the HA and NA subtypes. Two glycosites were located at HA0 cleavage sites and fusion peptides and were strikingly conserved in all HA subtypes, while the remaining glycosites were unique to their subtypes. Two to four conserved glycosites were found in the stalk domain of NA, but these are affected by the deletion of specific stalk domain sequences. Another highly conserved glycosite appeared at the top center of tetrameric global domain, while the others glycosites were distributed around the global domain. Here we present a detailed investigation of the distribution and the evolutionary pattern of the glycosites in the envelope glycoproteins of IVs, and further focus on the H5N1 virus and conclude that the glycosites in H5N1 have become more complicated in HA and less influential in NA in the last five years.