Mutations create challenges for typing Influenza A viruses

From BCCDC:

Today the Canadian Sentinel Practitioner Surveillance Network (SPSN) published a paper in Eurosurveillance, entitled “Mutations acquired during cell culture isolation may affect antigenic characterization of influenza A(H3N2) clade 3C.2a viruses”.

The paper is available in full here (open access). We thought you would be interested in the findings, although like influenza itself, they are complicated. We have summarized the key findings below but please consult the full paper for details and if you have any questions, please don’t hesitate to ask.

Study rationale

Influenza A(H3N2) viruses belonging to clade (or genetic group) 3C.2a were the main cause of the epidemic during the 2014/15 season and continue to predominate among H3N2 detections during the current 2015/16 season.

Globally, laboratories have had difficulty characterising H3N2 clade 3C.2a viruses antigenically by the conventional haemagglutination inhibition (HI) assay in order to assess vaccine relatedness (or match). The HI assay method relies on the ability of viruses to agglutinate red blood cells (RBCs) and many recently circulating H3N2 viruses have shown variable agglutination of RBCs, or have lost the ability to agglutinate RBCs. This is a particular problem for H3N2 clade 3C.2a viruses.

For H3N2 viruses that cannot be characterised by HI assay, laboratories have imputed antigenic relatedness based on sequencing findings, assuming that viruses that could be antigenically characterised within a given genetic clade are broadly representative of circulating strains. In this paper, we assessed that assumption.

Study synopsis

H3N2 clade 3C.2a viruses circulating in nature have a threonine (T) amino acid at position 160 of the haemagglutinin surface protein. Together with the amino acid asparagine (N) at position 158 and tyrosine (Y) at position 159, this N-Y-T sequon at positions 158-160 is a clade-defining feature that confers a potential gain of glycosylation to the haemagglutinin protein of clade 3C.2a viruses. At this location, the glycosylation motif can mask pivotal viral epitopes at the top of the haemagglutinin protein, potentially relevant to antibody recognition and binding.

In this paper, we show that the small proportion of H3N2 clade 3C.2a viruses that could be characterised by HI assay were not representative of the viruses circulating in nature with respect to the presence of this glycosylation motif. We found that the glycosylation motif had been modified in viruses that could be characterised antigenically by HI assay, an observation also previously noted by the European Centre for Disease Prevention and Control (ECDC). Furthermore, we show that loss (or partial loss) of the glycosylation motif is an artefact of growing the viruses in cell culture – a process required by labs before characterisation by HI assay.

Our findings suggest that caution is warranted in extrapolating antigenic relatedness or anticipating vaccine performance based on limited HI results for H3N2 clade 3C.2a viruses that continue to circulate globally. Our findings also underscore that, in addition to laboratory monitoring of viral genomic and antigenic relatedness to vaccine, ongoing epidemiological monitoring of vaccine effectiveness is important.