Antigenic Patterns and Evolution of the Human Influenza A (H1N1) Virus

Abstract

The influenza A (H1N1) virus causes seasonal epidemics that result in severe illnesses and deaths almost every year. A deep understanding of the antigenic patterns and evolution of human influenza A (H1N1) virus is extremely important for its effective surveillance and prevention. Through development of antigenicity inference method for human influenza A (H1N1), named PREDAC-H1, we systematically mapped the antigenic patterns and evolution of the human influenza A (H1N1) virus. Eight dominant antigenic clusters have been inferred for seasonal H1N1 viruses since 1977, which demonstrated sequential replacements over time with a similar pattern in Asia, Europe and North America. Among them, six clusters emerged first in Asia. As for China, three of the eight antigenic clusters were detected in South China earlier than in North China, indicating the leading role of South China in H1N1 transmission. The comprehensive view of the antigenic evolution of human influenza A (H1N1) virus can help formulate better strategy for its prevention and control.

Figure 1. Methodology and validation of the PREDAC-H1 method.

(a) Workflow of the PREDAC-H1 method, antigenic correlation network was illustrated using Cytoscape software. (b) Retrospective testing to infer the antigenic relationships between influenza A (H1N1) viruses. (c) Comparison of inferred antigenic clusters and predominant clusters in the US from the 1994–1995 to 2013–2014 epidemic seasons as reported by US CDC. Five dominant clusters are colored and labeled.

Figure 2. Antigenic and genetic evolution of human influenza A (H1N1) viruses.

(a) Phylogenetic tree of the HA1 region of the H1N1 HA protein. The tree was rooted using A/Brevig Mission/1/1918 strain as outgroup. (b) Inferred antigenic correlation network and antigenic clusters of human influenza A (H1N1) viruses. Clusters are colored and named as the abbreviation of vaccine strains. The circulation periods of each cluster are provided in parenthesis. (c) Genetic distance (calculated as amino acid substitutions) on HA1 of human influenza A (H1N1) strains to the A/Brevig Mission/1/1918 (H1N1) virus, each dot represents a virus strain. The strains were colored as antigenic clusters in Fig. 2b. The strains from 1918 to 1957, strains from 1977 to 2008, the swine-origin human-infecting strains were separated by gray lines. (d) Genetic distance (calculated as amino acid substitutions) on HA1 of human influenza A (H3N2) strains to the A/Hong Kong/1/1968 (H3N2) virus. Strains were colored as the antigenic clusters they belonged to.

Figure 3. Comparison of antigenic patterns in different regions from 1980 to 2013.

Dynamic changes in the percentage of antigenic clusters in Asia, Europe and North America were recorded yearly. The earliest appearance of a new antigenic cluster is marked by a colored bar.

Figure 4. Antigenic patterns in South and North China.

(a) Geographic distribution of H1N1 HA sequences provided by the China CDC and obtained from a public database from 1977 to 2008. The map was reconstructed using OpenStreetMap (http://www.openstreetmap.org/) with further modification, and is for illustrative purposes only. (b) Dynamic changes in the percentage of antigenic clusters in South and North China. The underlying bar represents the circulation period of each antigenic cluster in South and North China and the earliest appearance of a new antigenic cluster is marked with a star. (c) Co-circulation of antigenic clusters in South and North China. Co-circulation was measured as the entropy of co-circulating antigenic clusters.