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-NSS061–063 and NNS160–162 glycosylation sites were prevalent among the 2011–2012 Guangdong strains but the NA-NRS402–404 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.


Introduction
Influenza virus undergoes rapid evolution by both antigenic shift and antigenic drift and this presents a significant challenge for vaccine design to best match with viruses likely to circulate in the coming influenza season [1,2]. Since their emergence in 1968, influenza H3N2 viruses tend to be highly prevalent most years. The two dominant proteins in this particular virus strains are the H3 hemagglutinin (HA) and N2 neuraminidase (NA) surface glycoproteins [1,3]. Antigenic drift, also known as the result of positive selection, generally involves both of these proteins, resulting in evasion of the host immune response [4]. Many amino acid variations occur at the antibody binding sites, creating diversity in the HA and NA proteins and allowing them to evade host antibodies. However, these variations do not significantly alter the stereochemical or functional properties of HA and NA.
In  [5]. Antigenic and molecular characterization of H3N2 viruses over those three seasons revealed that the number of HA mutations is important, along with the nature and location of key mutations, and likely plays a significant role in antigenic drift. In previous work, amino acid substitutions were seen in five antigenic regions of HA1 from 2007-2011 Guangdong (for short GD) isolates; in particular, regions B (N160K) and D (K174R/ N). The K189E/N/Q and T228A mutations within the receptor-binding site (RBS) were present in the 2010 strains, affecting the antigenicity of HA1 [6]. The antigenicity of epidemic H3N2 isolates in 2011 differed from that of the A/Perth/16/2009 strain.
The evolution and epidemiology of influenza H3N2 viruses is partially related to global migration dynamics [7]. The evolution of H3N2 influenza over the past 10 years reflects the dynamics of a global metapopulation. According to the Center for Public Health Surveillance and Information Service of China [8], the number of influenza cases was 2.35-fold greater for January-June 2012 (74,151 cases) than for January-June 2011 (31,551 cases). Approximately 87% (365/418) of isolates were of the H3N2 subtype; these isolates were sourced from local epidemics and sporadic cases reported in Guangdong Province from March-June 2012. We analyzed HA and NA gene sequences of H3N2 viruses isolated from July 2011 to July 2012.

Viruses and genes
With assistance from the Guangdong Influenza Surveillance Network, H3N2 viruses from 2011-2012 were isolated in MDCK cell lines [6]. Viruses investigated in our study included four 2011 and four 2012 isolates (GenBank Accession Nos. CY125677-CY125692). Isolates from Zhuhai, Shaoguan, Foshan, Maoming and Meizhou are presented in Figure 1 and Table 1 16 HA genes from Asia (4 genes from Japan and Russia Asia region), Europe (8 genes from Russia and George) and North America (4 genes from USA) and 9 NA genes from Asia (2 genes from Russia Asia region) and Europe (7 genes from Russia and George) isolated in 2012.

Molecular detection of viral genes
Primers specific for the HA and NA genes of human H3N2 isolated from 1968-2010 were designed using Primer Premier 5.0 (Premier Biosoft International, Palo Alto, CA, USA) and synthesized by Life Technologies (Shanghai, China). These primers have been used in a previous study [6]. Viral RNA was extracted using a Qiagen QIAamp Viral RNA mini Kit, and reverse transcription polymerase chain reactions (RT-PCRs) conducted with Qiagen Sensiscript Reverse Transcriptase and Takara Pyr-oBest Taq. Amplicons were purified with a Qiagen Gel Extraction Kit and sequenced with an ABI PRISM BigDye Terminator v3.0 Ready Reaction Cycle Sequence Kit on an ABI PRISM 3100 Genetic Analyzer. The obtained sequences were analyzed using Lasergene 7.1.

Phylogenetic analysis and evolutionary dynamics
Virus population dynamics over time were estimated using the BEAST package v1.70 [9] and applying Bayesian Markov chain Monte Carlo (MCMC) methods. Phylogenetic trees were generated, and the reliability of trees tested by bootstrap analysis using 1000 replicates. The MCMC analysis was run for 50,000,000 generations, with stationarity and mixing efficiency examined using Tracer.

Evolutionary selection and N-glycosylation
Single likelihood ancestor counting (SLAC) was appropriate for large alignments but possibly underestimates the number of positively selected sites [10]. The fixedeffects likelihood (FEL) and the internal fixed-effects likelihood (IFEL) took the synonymous and non-synonymous substitutions into account and could be efficiently parallelized. Potential N-glycosylation sites were predicted using NetNGlyc 1.0 server [11]. This software application predicts N-glycosylation sites in human proteins using artificial neural networks that examine the N-X-S/T (Asn-Xaa-Ser/Thr) amino acid sequence [12].

Antigenic analysis, epitope region and three-dimensional (3D) structure
Cross-reactivity of isolated viruses was investigated using hemagglutination inhibition (HI) assays [13]. The variation sites in H3N2 virus HA and NA gene sequences isolated from 2011 and 2012 were compared with previously identified epitopes of HA and NA proteins [14]. The 3D structures of HA and NA proteins were established using the I-TASSER server [15]. Models were modified with UCSF Chimera1.5.3 [16].  We   Table 3). The 3D structures of HA and NA proteins for A/ GD/1154/2012 were modeled; HA (Q49R, A214S, V239I, N328S and D503N) and NA (L81P, D/N93G, S367N, K369N and N402D) variation sites were labeled in the 3D structures ( Figure 3).

Discussion
Mutated amino acids around human leukocyte antigen (HLA)-associated sites, especially those that are typically conserved, suggest that cooperative interactions act to preserve the local structural stability and protein function when mutations occur that confer evasion of cytotoxic T lymphocytes (CTLs) [17]. For the HA genes of isolated strains, the D69N, Y110H, I246V and E296A/T substitutions occurred around 2010 [6], while A214S, V239I and N328S have been present since 2011. Accumulation of HA (D69N, Y110H, I246V, E296A/T, A214S, V239I and N328S), along with NA (L81P and D93G) mutations, drove antigenic drift possibly giving rise to the H3N2 influenza epidemic in Guangdong and Southern China in 2012. The HA-A214S, -V239I and -N328S mutations occurred in the second half of 2011 (Table 1 and Figure 2). As the 2012 influenza season approached, these variations in circulating strains became prevalent, resulting in local epidemics throughout Southern China.
According to a previous study that focused on isolates from 2011 [18], HA1 genes were sub-divided among the P (A/Perth/16/2009-Clade) and V (A/Victoria/208/2009-Clade) clades. The former included subgroups1 and 2, while the latter included subgroups 3-6. When we compared our findings with the results from Klimov [18] and Huang [6], mutations A/N144T/D and K/N145N/S were only observed in the previous studies; the mutations we have reported here were not evident in Klimov's work. The mutations in the 2012 Guangdong isolates appear to have given rise to antigenic drift. Generally, five HA1 antigenic regions are analyzed and associated with antigenic drift that results in an epidemic. However, for the particular epidemic we investigated, these five regions were not seemingly associated with antigenic drift. Only the A214S substitution occurred in epitope D and the RBS. Beside the five HA1 antigenic regions, many amino acid sites were unclassified but still considered important [6]. Using the H3N2 HA and NA protein sequences of A/New York/ 348/ 2005 as a reference, all mutated amino acid sites were related to previously identified epitopes, which might involve in antigenic presentation / recognition / response. This indicated that mutations of B cell epitope / T cell epitope regions in the Immune Database influenced antigenicity to some degree [14]. These B cell epitopes / T cell epitopes span 17-18 amino acids in the A/New York/348/ 2005 strain. Further research is required to confirm the relationship between antigenicity and mutations in HA and NA genes.
As was observed for of major capsid L1 protein of HPV- 16    indicating that these sites avoided immunological pressure for their continued persistence [19]. Only 76% of positive scored sequons according to NetN-Glyc are modified by N-Glycans with a bias towards Thrcontaining sequons [20]. Compared with A/Perth/16/2009, two glycosylation sites (NSS 061-063 and NNS 160-162 ) in HA genes were prevalent in 2011-2012 Guangdong strains, with the NA glycosylation site N 402 D deleted. There are five possible functions for glycosite migration in human influenza viruses [21]: to more effectively mask the antigenic sites; to more effectively protect enzymatic cleavage sites of NA; to stabilize polymeric structures; to regulate receptor binding and catalytic activities; and to balance the binding activity of HA with the release activity of NA. Five HA glycosites and two NA glycosites were positive sites in this study. Gains in N-glycosylation sites were likely to be positively selected for shielding antigenic sites from immune responses [22]. The acquisition of glycosylation at residues 144 of pH1N1 was associated with viral replication, virulence and transmissibility and provided insights into the evolution dynamics of influenza viruses with implications in vaccine immunogenicity [23]. The loss of a glycosylation site might also be important in antibody recognition [22], where a deletion of the NA-NRS 402-404 occurred in the 2011-2012 Guangdong and other isolates in this study. With respect to HA/NA N-glycosylation changing, further analysis is required to determine any relationship between their structure mutation and function influence. At the antigenic sites of HA protein, positive selection appeared to have effected radical and conservation substitution in term of the charge of the amino acids, suggesting that antigenic drift is not a byproduct of HA evolution in H3N2 viruses [22].
Kenyan H3N2 viruses isolated during 2006-2007 revealed unique genetic variations, with several amino acid substitutions located at immunodominant epitopes of the HA1 protein [24]. These mutations included V112I at site E, K173 E at site D and N278K at site C. These mutations possibly result in a conformational change to the HA molecule, thereby exposing novel epitopes and thus abrogating the binding of pre-existing antibodies at these sites. A Canadian study in 2011 involving antigenic and molecular characterization of H3N2 viruses over three seasons revealed that the number of HA mutations was important, along with the nature and location of key mutations, and played a significant role in antigenic drift [5]. From our findings, presented in this report, the 2012 strains had evolved genetically and antigenically from the A/Perth/16/2009 vaccine-like strains. The A/GD/ 1154/2012 strain was antigenically distinct from the A/ GD/1105/2009 strain, suggesting that it may be the parental strain responsible for the 2012 H3N2 influenza virus epidemics in Southern China. Although influenza H3N2 viruses varied genetically and antigenically during 2009-2011 [6], we concluded that antigenic drift in 2012 along with the accumulation of evolutionary mutations in viruses isolated in 2012 played an important role in the influenza epidemics of Guangdong and Southern China during that period.