Abstract
The extensive circulation of Highly Pathogenic (HP) H5N1 Avian Influenza in Egypt in poultry since 2006 resulted in the emergence of distinct clades with the recent identification of a further clade: 2.2.1.1. The aim of this study was to characterize for the first time the antigenic profile of an extensive collection of genetically diverse Egyptian H5N1 HP viruses isolated between 2007 and 2010 applying antigenic cartography and principal component analysis to serological data. We identified that Egyptian H5N1 viruses have undergone significant antigenic diversification between 2007 and 2010 and two distinct antigenic clusters co-circulated in 2010. Such clusters correlated with 2.2.1 and 2.2.1.1 clades, showing for the first time that the new emerging clade 2.2.1.1 is antigenically distinct. This study highlights that the antigenic diversity of H5N1 HP Egyptian viruses may represent a potential challenge for the development of an effective vaccination programme for animal and human health in Egypt.
Keywords: H5N1 HPAI, Egypt, Antigenic drift, Antigenic cartography, PCA
Introduction
Highly pathogenic avian influenza (HPAI) viruses of the H5N1 subtype descending from A/goose/Guangdong/1/96 lineage were first detected in Africa in 2006 (Chen et al., 2006). Egypt was the second African country affected by H5N1 HPAI, which reported cases of infection in poultry in early 2006. Initial attempts to control the epidemics by implementing stamping out and movement control measures were largely unsuccessful.
In March 2006 vaccination of poultry was authorized essentially using inactivated vaccines derived from low pathogenic H5N2 Mexican or H5N1 Asian strains (Peyre et al., 2009). Notwithstanding vaccination and other control measures applied, H5N1 HPAI infection continued to spread and became endemic in Egypt in 2008.
To June 7, 2012, H5N1 HPAI virus caused 168 human cases, of which 60 were fatal (http://www.who.int/influenza/human_animal_interface/avian_influenza/archive/en/index.html).
Clade 2.2.1 was introduced into Egypt and spread rapidly in commercial and backyard flocks (W H O/OIE/FAO, 2009). As a result of the persistence and extensive circulation of H5N1 HPAI viruses in Egypt, variant strains emerged evolving into distinct genetic sub-clades (Cattoli et al., 2011a). Cattoli et al. (2011a) showed that the majority of H5N1 HPAI viruses isolated between 2007 and 2010 fell within 2 main genetic groups, already identified by Balish et al. (Balish et al., 2010) on a smaller data set, and identified as A and B subclades. Amino acid substitutions in the receptor binding domain (RBD), presence of distinct potential glycosylation sites, and differences in evolutionary rates were the characteristics that differentiated the 2 genetic sub-clades that circulated between 2007 and 2010 in Egypt. Recently the WHO has identified 12 new H5N1 clades and the Egyptian sub-clade 2.2.1 was further split into a new sub-clade 2.2.1.1, corresponding to genetic subclade B, indicating further divergence of contemporary strains of H5N1 circulating in Egyptian poultry (WHO/OIE/FAO H5N1 Evolution Working Group, 2012).
Our earlier study showed that an Egyptian H5N1 HPAI virus from 2008 exhibited a low cross reactivity in haemagglutination-inhibition (HI) tests against the Mexican vaccine seed strain (H5N2) commonly used in Egypt, suggesting that significant antigenic drift occurred (Cattoli et al., 2011b). In this study, mutations at residues 74, 141, 140, 144, 162 (H5 numbering) in the RBD of a 2008 isolate (A/chicken/Egypt/1709-6/2008) were identified as being responsible for the antigenic drift (Cattoli et al., 2011b). However, there is still lack of systematic study of antigenic properties of H5N1 HPAI Egyptian viruses in fact previous studies were based on the characterization of few isolates which were isolated before 2010 (Balish et al., 2010; Grund et al., 2011). The goal of this study was to evaluate for the first time the antigenic characteristics of an extensive and representative collection of recent Egyptian H5N1 HPAI viruses by the antigenic cartography and principal component analysis (PCA) and to assess the antigenic differences of clade 2.2.1 and 2.2.1.1 viruses that circulated in Egypt in 2007 and 2010. Such systematic study will provide insights for development of more effective vaccination programme in Egypt.
Results
Genetic characterization of Egyptian H5N1 viruses
Analysis of the haemagglutinin (HA) phylogenetic tree identified that viruses under study fall within the clade 2.2.1 or 2.2.1.1 (Fig. 1) recently identified by the WHO (WHO/OIE/FAO H5N1 Evolution Working Group, 2012). The H5N1 HPAI Egyptian viruses used in this study and isolated in 2007 belong to clade 2.2.1 (Fig. 1), viruses isolated in 2008 to clade 2.2.1.1 while viruses isolated in 2010 belong to both sub-clades: 2.2.1 and 2.2.1.1 (Fig. 1). Twenty-six H5N1 HPAI viruses located through phylogenetic trees (marked in colour) were selected as prototype viruses for antigenic characterization, and such a selection would expect to represent a systematic picture for the antigenic profile of Egyptian H5N1 HPAI viruses.
Fig. 1.
Phylogenetic tree of the amino acid sequences of the HA gene of the Egyptian H5N1 HPAI viruses constructed by the neighbor-joining method. The viruses included in the antigenic study are coloured according to the antigenic group they belong to. The numbers at each branch point represent bootstrap values and they were determined by bootstrap analysis using 1000 replications. Scale bar=0.04 amino acid substitutions/site.
Antigenic characterization of Egyptian H5N1 viruses
Antigenic cartography demonstrated for the first time that Egyptian H5N1 HPAI viruses can be intuitively separated into two major antigenic clusters EG-antigen-A and EG-antigen-B, which corresponded to clade 2.2.1 and 2.2.1.1 respectively. Specifically, the tested H5N1 HPAI viruses: 1709-1/07, 1709-2/08, 1709-4/07, 1709-10/08, 4337-343/08, 1553-1/10, 1553-15/10, 3982-5/10, 3982-8/10 and 3982-9/10, all of which belong to genetic clade 2.2.1, formed cluster EG-antigen-A (Fig. 2, Tables 1 and S1–2). The other 15 Egyptian H5N1 HPAI viruses tested formed cluster EG-antigen-B, and with the exception of strain 3982-44/2010 belonged to clade 2.2.1.1. Strain 3982-44 was shown to belong to cluster EG-Antigen-B but genetically grouped with clade 2.2.1. The average antigenic distances in EG-antigen-A and -B were 1.23 (standard deviation, 0.37) and 1.56 (standard deviation, 0.73) units, respectively, and each unit corresponds to 2 log2 HI titre (Table S2). The maximum distance between EG-antigen-A and EG-antigen-B was 5.06 units (between strains 1553-1/2010 and 1553-26/2010), the minimum distance between EG-antigen-A and EG-antigen-B was 1.63 units (between 1709-1/2008 and 3982-19/2010), and the average distance between EG-antigen-A and EG-antigen-B was 3.57 units (Table S2). A student t-test showed that the antigenic distances within each antigenic cluster was significantly different than those between the two identified antigenic clusters by hierarchical clustering (p < 0.001).
Fig. 2.
Influenza antigenic cartography for Egyptian H5N1 HPAI viruses from 2007 to 2010. Viruses in light blue belong to EG-antigen-A1 isolated in 2008, viruses in dark blue belong to EG-antigen-A2 isolated in 2010, viruses in yellow belong to EG-antigen-B1 group and viruses in red belong to EG-antigen-B2 group. In purple is virus A/chicken/Egypt/3982-44/2010, in black virus A/chicken/Nigeria/4337-343/2008 and in green virus A/chicken/Mexico/1994 (H5N2).
Table 1.
List of viruses under study, their abbreviations, genetic clade and the antigenic group. Viruses are coloured according to colours in the antigenic map./: no information available.
| Viruses | Abbreviations | Antigenic group (EG antigen) | Genetic clade | Area | Period of detection |
|---|---|---|---|---|---|
| A/CHICKEN/EGYPT/1709-1/2007(H5N1) | 1709-1/07 | A1 | 2.2.1 | / | 25/02/2007 |
| A/CHICKEN/EGYPT/1709-2/2008(H5N1) | 1709-2/08 | A1 | 2.2.1 | / | / |
| A/CHICKEN/EGYPT/1709-4/2008(H5N1) | 1709-4/07 | A1 | 2.2.1 | / | 04/03/2007 |
| A/DUCK/EGYPT/1709-10/2008(H5N1) | 1709-10/08 | A1 | 2.2.1 | / | / |
| A/CHICKEN/NIGERIA/4337–343/2008 | 4337–343/08 | A1 | 2.2.1 | / | / |
| A/CHICKEN/EGYPT/1553-1/2010(H5N1) | 1553-1/10 | A2 | 2.2.1 | Domiat | 02/2010 |
| A/CHICKEN/EGYPT/1553-15/2010(H5N1) | 1553-15/10 | A2 | 2.2.1 | Cairo | 01/2010 |
| A/CHICKEN/EGYPT/3982-5/2010(H5N1) | 3982-5/10 | A2 | 2.2.1 | Behera | 11/04/2010 |
| A/BROILER/EGYPT/3982-8/2010(H5N1) | 3982-8/10 | A2 | 2.2.1 | Alex | 20/02/2010 |
| A/BROILER/EGYPT/3982-9/2010(H5N1) | 3982-9/10 | A2 | 2.2.1 | Alex | 20/02/2010 |
| A/CHICKEN/MEXICO/232/94 | Mexico | - | / | / | |
| A/CHICKEN/EGYPT/1709-5/2008(H5N1) | 1709-5/08 | B1 | 2.2.1.1 | / | 08/01/2008 |
| A/CHICKEN/EGYPT/1709-6/2008(H5N1) | 1709-6/08 | B1 | 2.2.1.1 | / | 03/01/2008 |
| A/CHICKEN/EGYPT/3982-19/2010(H5N1) | 3982-19/10 | B1 | 2.2.1.1 | / | 02/07/2010 |
| A/CHICKEN/EGYPT/1553-2/2010(H5N1) | 1553-2/10 | B2 | 2.2.1.1 | Domiat | 02/2010 |
| A/CHICKEN/EGYPT/1553-6/2010(H5N1) | 1553-6/10 | B2 | 2.2.1.1 | Dkahlia | 02/2010 |
| A/CHICKEN/EGYPT/1553-28/2010(H5N1) | 1553-28/10 | B2 | 2.2.1.1 | Sharqia | 02/2010 |
| A/BROILER/EGYPT/3982-3/2010(H5N1) | 3982-3/10 | B2 | 2.2.1.1 | Behera | 01/02/2010 |
| A/CHICKEN/EGYPT/3982-50/2010(H5N1) | 3982-50/10 | B2 | 2.2.1.1 | / | / |
| A/CHICKEN/EGYPT/3982-52/2010(H5N1) | 3982-52/10 | B2 | 2.2.1.1 | Dakahlia | / |
| A/CHICKEN/EGYPT/2095-75/2010(H5N1) | 2095-75/10 | B2 | 2.2.1.1 | El Behera | 27/03/2010 |
| A/BROILER/EGYPT/3982-21/2010(H5N1) | 3982-21/10 | B2 | 2.2.1.1 | / | 02/07/2010 |
| A/CHICKEN/EGYPT/1553-13/2010(H5N1) | 1553-13/10 | B2 | 2.2.1.1 | Cairo | 01/2010 |
| A/CHICKEN/EGYPT/1553-26/2010(H5N1) | 1553-26/10 | B2 | 2.2.1.1 | Cairo | 01/2010 |
| A/CHICKEN/EGYPT/3982-78/2010(H5N1) | 3982-78/10 | B2 | 2.2.1.1 | Sharkia | 02/02/2010 |
| A/CHICKEN/EGYPT/3982-44/2010(H5N1) | 3982-44/10 | B2 | 2.2.1 | Dakahlia | / |
Based on genetic data, we additionally performed the Principal Component Analysis (PCA) analysis in order to assess the antigenic evolution and trajectories. PCA showed that the Egyptian viruses tested also formed two antigenic groups confirming the antigenic cartography data, EG-antigen-A and EG-antigen-B, and that these groups could be further separated into subgroups: EG-antigen-A1 and –A2, and EG-antigen-B1 and B2 respectively (Figs. S1–S3). The viruses in each antigenic subgroups are listed in Table 1: EG-antigen-A1 comprised viruses isolated in 2007 and 2008 (1709-1/2007; 1709-2/2008; 1709-4/2007; and 1709-10/2008; 4337-343/07) and EG-antigen-A2 comprised five viruses isolated in 2010 (1553-1/2010, 1553-13/2010, 3982-5/2010, 3982-8/2010, 3982-9/2010); EG-antigen-B1 comprised 2 viruses isolated in 2008 (strain 1709-5/ 2008 and 1709-6/2008) and one virus isolated in 2010 (strain 3982-19/2010), and EG-antigen-B2 comprised viruses isolated only in 2010 (Table 1, Figs. 2 and S2). The average distances and mean values of each of these subgroups are shown in Table S2.
The MN assay confirmed the HI data and showed that anti group A sera did not neutralize group B viruses and vice versa. The MN assay also confirmed that the 3982-44/10 virus does not belong to EG-antigen-A because no neutralization was observed with homologous antisera to this group (Table S3).
Antigenic cartography demonstrated that the vaccine strain A/chicken/Mexico/1994 (H5N2) was antigenically distinct from the H5N1 Egyptian viruses, including both antigenic clusters EG-antigen-A and -B. However, A/chicken/Mexico/1994 (H5N2) was antigenically closer to the EG-antigen-A than the EG-antigen-B (Fig. 2, Table S2).
A/chicken/Nigeria/4337–343/2008, the earlier strain introduced into Africa was located in antigenic cluster A, which included most of the 2008 isolates. This observation was validated with the HI assay. With reference to the spatial distribution of viruses (Fig. 3 and Table 1), available information is fragmentary and precluded making conclusions on the prevalence of EG-antigen-A or B viruses in Egypt. Instead, both EG-antigen-A and B were co-circulating at the same time in Egypt, e.g. in year 2008, and 2010. The geographic distributions of EG-antigen-A or B did not show relevant associations between geographic locations and antigenic clusters, either (Fig. 3).
Fig. 3.
Geographical location of H5N1 HPAI antigenic clusters in Egypt.
Molecular characterization of antigenic groups
A total of 20 amino acid (aa) differences were observed in the antigenic sites of the HA gene of viruses under study. Seven of the differences were detected in the antigenic site A, 9 in the antigenic site B, 1 in the antigenic site D, 1 in the antigenic site E and 2 falling outside these antigenic sites (Table 2). Thirteen of these amino acid mutations were observed to have different prevalent patterns among the antigenic subgroups identified through antigenic cartography and included residues 151 (155, H3 numbering, antibody binding site B in H3), 154 (158, B), 156 (160, B), 162 (166, B), 190 (194, B), 120 (125, A), 129 (133, A), 140 (144, A), 141 (145, A), 226 (230, D), 74 (82, E), 97 (104), and 144 (148) (Table 2). The aa mutation, observed at position 144 near the 130 loop, was involved in the vaccine drift of 1709-6/2008 strain (Cattoli et al., 2011b). Positions 120, 140, 141, 154, 156 were confirmed to affect the antigenicity of H5N1 HPAI viruses (Cai et al., 2010). The majority of antigenic cluster EG-Antigen-B (clade 2.2.1.1) viruses presented the following aa mutations against EG-Antigen-A: D97N, P74S, F144Y, S141P, M226V and A156T. The differences between EG-Antigen-A1 and EG-Antigen-A2 can be found in residues 120, 151 and 154. The differences between EG-Antigen-B1 and EG-Antigen-B2 were less apparent. The majority of viruses belonging to the antigenic clusters B1 and B2 showed differences in residues 129, 154, 156, 162, and 190. This indicated the complication of antigenic diversity in Egyptian H5N1 viruses. Further experiments are required to confirm the role of each residue in affecting antigenic properties for these Egyptian H5N1 viruses.
Table 2.
Amino acid changes in the HA molecule of H5N1 HPAI Egyptian viruses under study. The H5 and H3 numbering are given. Viruses are listed according to the antigenic group and genetic clade they belong to -:deletion.
| H5 position | 151 | 154 | 156 | 162 | 190 | 120 | 129 | 140 | 141 | 226 | 97 | 74 | 144 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
|
||||||||||||||
| H3 (antibody binding site) | 155 (B) | 158 (B) | 160 (B) | 166 (B) | 194 (B) | 125 (A) | 133 (A) | 144 (A) | 145 (A) | 230 (D) | 104 | 82 (E) | 148 | ||
|
|
|
||||||||||||||
| Viruses | Antigenic group | Clade | |||||||||||||
|
| |||||||||||||||
| A/CHICKEN/EGYPT/1709-1/2007 | A1 | 2.2.1 | I | D | A | R | L | N | S | R | S | M | N | P | F |
| A/CHICKEN/EGYPT/1709-2/2008 | A1 | 2.2.1 | I | D | A | I | L | S | S | R | S | M | D | P | F |
| A/CHICKEN/EGYPT/1709-4/2007 | A1 | 2.2.1 | I | D | A | R | L | S | S | R | S | I | D | P | F |
| A/DUCK/EGYPT/1709-10/2008 | A1 | 2.2.1 | I | D | A | K | L | S | S | R | S | I | D | P | F |
| A/CHICKEN/EGYPT/1553-1/2010 | A2 | 2.2.1 | T | D | A | R | L | N | – | R | S | M | D | P | F |
| A/CHICKEN/EGYPT/1553-15/2010 | A2 | 2.2.1 | T | N | A | K | L | G | – | R | S | M | D | - | F |
| A/CHICKEN/EGYPT/3982-5/2010 | A2 | 2.2.1 | T | N | A | K | L | N | – | R | P | M | D | P | F |
| A/BROILER/EGYPT/3982-8/2010 | A2 | 2.2.1 | T | N | A | K | L | N | – | R | S | M | D | P | F |
| A/BROILER/EGYPT/3982-9/2010 | A2 | 2.2.1 | T | N | A | K | L | N | – | R | S | M | D | P | F |
| A/CHICKEN/EGYPT/1709-5/2008 | B1 | 2.2.1.1 | I | D | A | K | L | S | S | G | P | V | N | S | Y |
| A/CHICKEN/EGYPT/1709-6/2008 | B1 | 2.2.1.1 | I | D | A | K | L | S | S | G | P | V | N | S | Y |
| A/CHICKEN/EGYPT/3982-19/2010 | B1 | 2.2.1.1 | I | N | T | K | I | N | L | G | L | V | N | S | Y |
| A/CHICKEN/EGYPT/1553-2/2010 | B2 | 2.2.1.1 | I | N | T | E | I | S | L | G | P | V | N | S | Y |
| A/CHICKEN/EGYPT/1553-6/2010 | B2 | 2.2.1.1 | I | N | T | E | I | S | L | E | P | V | N | S | Y |
| A/CHICKEN/EGYPT/1553-28/2010 | B2 | 2.2.1.1 | I | N | T | E | I | S | L | E | P | V | N | S | Y |
| A/BROILER/EGYPT/3982-3/2010 | B2 | 2.2.1.1 | I | N | T | E | I | S | L | E | P | V | N | S | Y |
| A/CHICKEN/EGYPT/3982-50/2010 | B2 | 2.2.1.1 | I | N | T | E | I | S | L | E | P | V | N | S | Y |
| A/CHICKEN/EGYPT/3982-52/2010 | B2 | 2.2.1.1 | I | N | T | E | I | S | L | G | P | V | N | S | Y |
| A/CHICKEN/EGYPT/2095-75/2010 | B2 | 2.2.1.1 | I | N | T | E | I | S | L | G | P | V | N | S | Y |
| A/BROILER/EGYPT/3982-21/2010 | B2 | 2.2.1.1 | I | N | T | K | I | S | L | G | P | V | N | S | Y |
| A/CHICKEN/EGYPT/1553-13/2010 | B2 | 2.2.1.1 | I | N | T | K | I | S | L | G | P | V | N | S | Y |
| A/CHICKEN/EGYPT/1553-26/2010 | B2 | 2.2.1.1 | I | N | T | E | I | S | L | G | P | V | N | S | Y |
| A/CHICKEN/EGYPT/3982-78/2010 | B2 | 2.2.1.1 | I | N | T | K | I | S | L | G | P | V | N | S | Y |
| A/CHICKEN/EGYPT/3982-44/2010 | B2 | 2.2.1 | T | D | A | K | L | N | – | R | S | M | D | P | F |
It is interesting that A/chicken/Egypt/3982-44/2010 (H5N1) has the residues similar to those in EG-Antigen-A, although antigenically this virus is more similar to those viruses located in EG-Antigen-B. Genetically this virus belongs to clade 2.1.1 as well. Further study is required to identify the unique features determining the antigenicity of this virus.
Discussion
In this study we used a serological approach to cluster H5N1 HPAI viruses isolated in Egypt between 2007 and 2010. The application of the antigenic cartography and PCA highlighted the similarity and diversity on a representative and wide-ranging collection of viruses. This study represents a first unique evidence of the existence of distinct antigenic clusters among Egyptian viruses circulating in 2010. In addition we were able to speculate on the evolution of antigenic clusters identified.
Through the antigenic cartography approach, in this study we have identified inter and intra clade antigenic differences of viruses belonging to clades 2.2.1 and 2.2.1.1. Inter and intra clade cross-reactivity were already studied for 9 clades including clade 2.2 (Ducatez et al., 2011).
Although each antigenic cluster was further divided in two sub-clusters, the average distance within each sub-cluster (A1–2 and B1–2) calculated trough the antigenic cartography approach is different. Viruses of EG-antigen-A1 and 2 have a minor antigenic distance (minimum and maximum) than viruses of EG-antigen-B1 and 2. The evidence that the 1709-6/2008 virus clusters in the EG-antigen-B1 group, together with 2010 isolates, confirmed a previous study (Cattoli et al., 2011b) that identified this virus as an antigenic variant compared to 1709-1/07 isolate (EG-antigen-A1).
There is one strain (A/chicken/Egypt/3982-44/2010), which behaved differently from an antigenic viewpoint. Although this virus belonged to 2.2.1 clade, it grouped with anti EG-antigen-B viruses by HI. However phylogenetic analysis of this virus did not show any peculiar aa mutations at the antigenic sites which could be responsible for such difference. The lack of detection of viruses similar in the antigenic and genetic properties to that one might be related to gaps in surveillance or they have been supplanted by the new clade 2.2.1.1. Performing studies on additional viruses may verify the hypothesis as to whether this virus represents a single case or a representative member of a different group.
In addition, PCA analysis suggested that from an antigenic point of view the EG-antigen-A2 and the EG-antigen-B clusters evolved from the EG-antigen-A1 viruses formed by earlier viruses and they may represent the antigenic ancestor viruses. Thus it seems that viruses followed two different antigenic evolutionary trajectories from 2007 to 2010; one trajectory resulted in the EG-antigen-A2 group and the other in the EG-antigen-B cluster. The EG-antigen B1 appeared in between of EG-antigen-A1 and B2 suggesting that already in 2008 the Egyptian H5N1 viruses were subjected to an antigenic differentiation.
Further the PCA showed that sera A/chicken/Egypt/1709-6/ 2008 cross-reacted well against both antigenic clusters. This was confirmed by the MN assay, and suggests that vaccine production may induce antibodies, which are cross-reactive and neutralizing in vitro against EG-antigen-A and B viruses.
The finding that the Mexican virus is more antigenically related to EG-antigen-A than B may indicate that the continued circulation in immunized and partially immunized poultry population of clade 2.2.1 was the driven force for their antigenic evolution as suggested by the higher antigenic distance of EG-antigen-B viruses than A, from the Mexican strain. However in Egypt several vaccine preparations are in use (Peyre et al., 2009) suggesting that the Egyptian viruses are not under a uniform immunological pressure. Whether the higher antigenic distance of the Mexican strain and the EG-antigen-B viruses may be indicative of reduced vaccine efficacy against this antigenic cluster, cannot be stated without performing in vivo tests. However, previous studies (Terregino et al., 2010) showed that the low cross reactivity of H5N2 Mexican strain with 2008 Egyptian strains did not predict the clinical and virological protection achieved following challenge with the 1709-6/2008 strain or genetically related viruses.
Although in vitro mutagenesis was not carried out to assess the real impact of aa changes observed between the 2 clades, most probably, as already shown (Cattoli et al., 2011b) the mutations in positions 74, 104, 140, 141 and 144 may be responsible for the antigenic differences of clades 2.2.1 and 2.2.1.1. As reported previously the P74S and R140G mutations in clade 2.2.1.1 virus resembled the antigenicity of a clade 2.2.1 virus suggesting the key role of this aa changes for the inter clade antigenic diversity (Cattoli et al., 2011b).
The M226V mutation was reported to occur in a unexposed part of the HA molecule and to not affect the antigenic profile of viruses 1709-1/07 (clade 2.2.1) and 1709-6/08 (clade 2.2.1.1): however, our study showed that this mutation is conserved among all clade 2.2.1.1 viruses under study suggesting this change may be advantageous for viral antigenic fitness of clade 2.2.1.1 strains. With reference to mutation N97D, which is present in all clade 2.2.1.1 viruses and not included in any antigenic site, we can speculate that such change may be the result of a vaccine pressure as for the P74S mutation (Cattoli et al., 2011b). This is suggested by the presence of D aa in position 97 in the Mexican vaccine seed strain (H5N2).
A conserved aa mutation at position 151 in the antigenic site B was detected between EG-antigen-A1 and 2 viruses which together with aa changes in position 120 (antigenic site A) may be responsible of this sub-clustering. Few aa mutations that differentiated EG-antigen-B1 and 2 viruses were detected most probably due to the low number (3) of EG-antigen-B1 viruses under study. However two of 3 EG-antigen-B1 viruses (1709-5/08 and 1709-6/08) shared the N154D, T156A (antigenic site B) and L129S (antigenic site A) aa mutations with EG-antigen-A1 and/or 2 viruses but not with EG-antigen-B2 viruses. These 2 viruses may represent an evolutionary snapshot from EG-antigen-A to B or from clade 2.2.1 to 2.2.1.1.
Conclusions
Through a systematic study of the antigenic profiling of Egyptian H5N1 HPAI viruses, we have shown for the first time that substantial antigenic differences exist inter and also intra clades. Our results demonstrated that multiple antigenic clusters co-circulated in Egypt in 2010, even in the same Egyptian province, as having generating the challenges in vaccination programme and pandemic preparedness. The antigenic evolution of the Egyptian H5N1 HPAI viruses followed a temporal pattern. We were also able to speculate on the impact of aa changes observed on the antigenic difference noticed and on the evolution of antigenic clusters. However further studies are necessary to elucidate the specific role of each aa mutations in the Egyptian H5N1 viruses. Data presented herein confirm that newly recognized clade 2.2.1.1 shows a distinct antigenic profile from 2.2.1 viruses. This study can be considered as first step for a more harmonized approach to study the antigenic characteristics of Egyptian H5N1 HPAI viruses. The results of this investigation emphasize the need to monitor constantly the genetic and antigenic evolution of H5N1 viruses in endemic countries such as Egypt in order to make educated decisions on vaccine candidates for humans and animals.
Materials and methods
Phylogenetic analysis
The phylogenetic analysis was used to select representative viruses to carry out the antigenic analysis and to identify mutations which may be involved in the antigenic diversification of viruses under study.
Sequences of the HA gene of fifty H5N1 HPAI viruses isolated in Egypt between 2007 and 2010, previously generated (Cattoli et al., 2011a), were used to construct a phylogenetic tree. Sequences of viruses under study were aligned and compared with some representative publicly available H5N1 HPAI sequences of viruses from Egypt.
The sequences were deposited into GenBank with accession numbers EU17849, EU17851, EU17853, EU17855, EU17857, CY047979, CY020645; CY016899; AY497096; and in the GISAID database under accession numbers: EPI81469–EPI81472; EPI81478–EPI81505; EPI20904; EPI156750; EPI156776; EPI156731; EPI287355; EPI81506–EPI81507; EPI20912.
For the HA gene, maximum likelihood (ML) tree was estimated using the best-fit general time-reversible (GTR)+I+Γ4 model of nucleotide substitution using PAUP* (Wilgenbusch and Swofford, 2003). Parameter values for the GTR substitution matrix, base composition, gamma distribution of the rate variation among sites (with four rate categories, Γ4), and proportion of invariant sites (I) were estimated directly from the data using MODELTEST V.3.7 (Posada and Crandall, 1998). A bootstrap resampling process (1000 replications) using the neighbor joining (NJ) method was used to assess the robustness of individual nodes of the phylogeny, incorporating the ML substitution model defined above.
Viruses included in the antigenic analysis
A total of 26 H5N1 HPAI viruses isolated from poultry were selected according to the phylogenetic analysis (Table 1). Four H5N1 viruses belonging to genetic clade 2.2.1 and two to clade 2.2.1.1 were isolated in 2008. The remaining H5N1 viruses were isolated during 2010: 6 belonged to clade 2.2.1 and 12 to clade 2.2.1.1. One H5N1 virus used in this study was isolated in 2007 (A/ chicken/Egypt/1709-1/2007) and belonged to clade 2.2.1. In addition, A/chicken/Mexico/1994 (H5N2) and A/chicken/Nigeria/4337-343/2008 (H5N1 HPAI clade 2.2) were included. A/chicken/Mexico/ 1994 (H5N2) was included as it is one of the vaccine strain widely used in commercial poultry in Egypt (Peyre et al., 2009).
Viruses were grown in 10–11 -day-old Specific Pathogen Free (SPF) chicken embryonating eggs and inactivated with 0.05% (v/v) of beta-propiolactone. Information on viruses (i.e. genetic clade, area of isolation) is listed in Table 1.
Antisera
Antisera were produced in SPF chickens using representative H5N1 HPAI viruses of genetic clades 2.2.1 and 2.2.1.1, isolated in 2008 and 2010. The following viruses were used to immunize chickens without adjuvant: A/chicken/Egypt/1553-1/2010, clade 2.2.1 (1553-1/10); A/chicken/Egypt/1553-15/2010, clade 2.2.1 (1553-15/10); A/chicken/Egypt/1553-2/2010, clade 2.2.1.1 (1553-2/10) and A/chicken/Egypt/1553-13/2010, clade 2.2.1.1 (1553-13/ 10). Groups of 10 chickens per virus were used. Five chickens were injected intravenously and five intramuscularly (without adjuvant) with 0.5 ml undiluted inactivated virus. Chickens were boosted once, 2 weeks after initial injection with 0.5 ml undiluted virus. A booster was necessary to achieve a minimum HI titre of 5log 2 to carry out HI tests. Inactivated viruses used to immunized birds showed haemagglutinating titres between 6 and 7log 2. Serum samples were collected 3 weeks after the second immunization. Blood was collected and stored per individual bird. Sera against viruses A/chicken/Egypt/1709-6/2008, clade 2.2.1.1 (1709-6/08) and A/chicken/Mexico/1994 (H5N2) were used as control and produced by intramuscular injection of 0.5 ml inactivated viral antigens emulsified with adjuvant (ISA 70VG, SEPPIC, France). A total of 43 chicken serum samples were collected but only 40 were used to carry out the cross HI tests as 3 chickens did not seroconvert. All animals were handled in strict accordance with the relevant national animal care guidelines.
Haemaglutination inhibition and microneutralization tests
The HI test was carried out according to OIE guidelines (OIE, 2010). Serum samples were tested individually, resulting in 40 HI titre values for each virus isolate. The antigenic dataset consisted of a table of 40 chicken sera by 26 viruses with 1040 individual HI measurements expressed as log 2.
The serum neutralizing activity was assessed by a standard microneutralization test (MN) assay performed as previously described (Rowe et al., 1999). Avian sera were treated with a receptor destroying enzyme (RDE) before use and heat inactivated at +56°C for 30 min. Anti 1709-6/08 (clade 2.2.1.1), anti Mexican H5N2, anti 1553-1/10 (clade 2.2.1) and anti 1553-13/10 (clade 2.2.1.1) sera were used in the MN assay to test their neutralizing activity against the homologous and 3982-44/10 viruses. The initial starting dilution of all sera was 1:80.
Antigenic cartography construction
The antigenic profiling of the 26 selected H5N1 HPAI viruses was analyzed using AntigenMap (http://sysbio.cvm.msstate.edu/AntigenMap). AntigenMap was developed for antigenic cartography construction. AntigenMap calculated antigenic distance based HI data by integrating low rank matrix completion and multiple dimensional scaling (Cai et al., 2010). The data normalization was performed as described elsewhere (Ducatez et al., 2011). The low rank matrix completion can not only help recover the missing values in HI data but also reduce noises in HI data. In this study, we utilized also the PCA, which derived the relationships among viruses from the HI data. PCA provided an additional method to confirm the antigenic relationship in antigenic cartography (see supplementary information).
Supplementary Material
Acknowledgments
Authors acknowledge Atem S Abd El-Hamid; Aly A. Hussein; Abdel Arafa, Mona Aly, Magdy Hassan for kindly providing viruses. Authors wish also to thank Francesco Bonfante, Serafino Pianta, Alessandro A. Leidi, Antonia Ricci and Dennis Senne for their support. Jiliang Yang and Xiu-Feng Wan were supported by NIH NIDAID RC1AI086830 and DOJ 2010-DD-BX-0596.
Appendix A. Supporting information
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.virol.2012.09.016.
Footnotes
Disclaimer
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the institutions with which the authors are affiliated.
References
- Balish AL, Davis CT, Saad MD, El-Sayed N, Esmat H, Tjaden JA, Earhart KC, Ahmed, Abd LE, El-Halem M, Ali AH, Nassif SA, El-Ebiary EA, Taha M, Aly MM, Arafa A, O’Neill E, Xiyan X, Cox NJ, Donis RO, Klimov AI. Antigenic and genetic diversity of highly pathogenic avian influenza A (H5N1) viruses isolated in Egypt. Avian Dis. 2010;54:329–334. doi: 10.1637/8903-042909-Reg.1. [DOI] [PubMed] [Google Scholar]
- Cai Z, Zhang T, Wan XF. A computational framework for influenza antigenic cartography. PLoS Comput Biol. 2010;6 (10):e1000949. doi: 10.1371/journal.pcbi.1000949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cattoli G, Fusaro A, Monne I, Coven F, Joannis T, El-Hamid HS, Hussein AA, Cornelius C, Amarin NM, Mancin M, Holmes EC, Capua I. Evidence for differing evolutionary dynamics of A/H5N1 viruses among countries applying or not applying avian influenza vaccination in poultry. Vaccine. 2011a;29(50):9368–9375. doi: 10.1016/j.vaccine.2011.09.127. [DOI] [PubMed] [Google Scholar]
- Cattoli G, Milani A, Temperton N, Zecchin B, Buratin A, Molesti E, Aly MM, Arafa A, Capua I. Antigenic drift in H5N1 avian influenza virus in poultry is driven by mutations in major antigenic sites of the hemagglutinin molecule analogous to those for human influenza virus. J Virol. 2011b;85:8718–8724. doi: 10.1128/JVI.02403-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chen H, Smith GJ, Li KS, Wang J, Fan XH, Rayner JM, Vijaykrishna D, Zhang JX, Zhang LJ, Guo CT, Cheung CL, Xu KM, Duan L, Huang K, Qin K, Leung YH, Wu WL, Lu HR, Chen Y, Xia NS, Naipospos TS, Yuen KY, Hassan SS, Bahri S, Nguyen TD, Webster RG, Peiris JS, Guan Y. Establishment of multiple sublineages of H5N1 influenza virus in Asia: implications for pandemic control. Proc Natl Acad Sci USA. 2006;103:2845–2850. doi: 10.1073/pnas.0511120103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ducatez MF, Cai Z, Peiris M, Guan Y, Ye Z, Wan XF, Webby RJ. Extent of antigenic cross-reactivity among highly pathogenic H5N1 influenza viruses. J Clin Microbiol. 2011;49:3531–3536. doi: 10.1128/JCM.01279-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Grund C, Abdelwhab EM, Arafa AS, Ziller M, Hassan MK, Aly MM, Hafez HM, Harder TC, Beer M. Highly pathogenic avian influenza virus H5N1 from Egypt escapes vaccine-induced immunity but confers clinical protection against a heterologous clade 2.2.1 Egyptian isolate. Vaccine. 2011;29:5567–5573. doi: 10.1016/j.vaccine.2011.01.006. [DOI] [PubMed] [Google Scholar]
- Peyre M, Samaha H, Makonnen YJ, Saad A, Abd-Elnabi A, Galal S, Ettel T, Dauphin G, Lubroth J, Roger F, Domenech J. Avian influenza vaccination in Egypt: limitations of the current strategy. J Mol Genet Med. 2009;3:198–204. doi: 10.4172/1747-0862.1000035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Posada D, Crandall KA. MODELTEST: testing the model of DNA substitution. Bioinformatics. 1998;14:817–818. doi: 10.1093/bioinformatics/14.9.817. [DOI] [PubMed] [Google Scholar]
- Rowe T, Abernathy RA, Hu-Primmer J, Thompson WW, Lu X, Lim W, Fukuda K, Cox NJ, Katz JM. Detection of antibody to avian influenza A (H5N1) virus in human serum by using a combination of serologic assays. J Clin Microbiol. 1999;37:937–943. doi: 10.1128/jcm.37.4.937-943.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Terregino C, Toffan A, Cilloni F, Monne I, Bertoli E, Castellanos L, Amarin N, Mancin M, Capua I. Evaluation of the protection induced by avian influenza vaccines containing a 1994 Mexican H5N2 LPAI seed strain against a 2008 Egyptian H5N1 HPAI virus belonging to clade 2.2.1 by means of serological and in vivo tests. Avian Pathol. 2010;39:215–222. doi: 10.1080/03079451003781858. [DOI] [PubMed] [Google Scholar]
- WHO/OIE/FAO H5N1 Evolution Working Group. Continued evolution of highly pathogenic avian influenza A (H5N1): updated nomenclature. Influenza Other Respi Viruses. 2012;6:1–5. doi: 10.1111/j.1750-2659.2011.00298.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wilgenbusch JC, Swofford D. Inferring evolutionary trees with PAUP*. Current Protocols in Bioinformatics. 2003;chapter 6(Unit 6.4) doi: 10.1002/0471250953.bi0604s00. [DOI] [PubMed] [Google Scholar]
- WHO/OIE/FAO, H5N1 evolution working group. Letter to the editor: continuing progress towards a unified nomenclature for the highly pathogenic H5N1 avian influenza viruses: divergence of clade 2.2 viruses. Influenza Other Respi Viruses. 2009;3:59–62. doi: 10.1111/j.1750-2659.2009.00078.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- OIE. Terrestrial Animal Health Code. 2010;Chapter 10.4 [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.