efore the pandemic. Therefore, we compared the epitopes of current seasonal viruses, which included the pandemic strains, with those of swine strains and high virulent strain H5N1 in order to find some implications in immune protection on human. The 96% (112/117) current seasonal epitopes were shared with swine viruses, as shown in Figure 4(a), suggesting that the human viruses have close relation with swine viruses (inter-host transmission or transmission by reassortant strains). If only considering the epitopes in general population, the same result was clearer (Figure 4(b)). All the epitopes in current seasonal strains were conserved in swine population. H5N1 shared much more epitopes with swine viruses than with human viruses and there was no epitope that only shared with human viruses. There were 28 epitopes shared among swine, H5N1 and human viruses. However, the viral populations in H5N1 and swine viruses contained many new epitopes to human, which should be a threat to human population by reassortant strains.
The swine viruses had more novel epitopes than those from humans due to the existence of more serotypes of swine strains. We thus separately compared the serotypes H1N1 and H3N2 from swine viruses as well as current seasonal human viruses, as shown in Figure 5. Only a few epitopes are specific to swine or to human population. The further comparison showed that all the epitopes specific to human strains or to swine strains in serotypes H1N1, H3N2 belonged to small populations (less than 30%) (not shown). The results indicated that the epitopes specific to human strains or to swine strains could be the consequence of viral mutations in each host. Of course, it is also possible that small proportional epitopes come from reassortant strains by sporadic infection on human.
Figure 3. Phylogeny analysis of 2009 pandemic H1N1 strain with the relation to other groups of strains based on the sequences of NA protein. The strain groups have been indicated in the figure.
Figure 4. Comparison of B-cell epitopes among current seasonal strains (including the novel pandemic strain), the swine strains and the highly virulent strain H5N1 since 1988. (a) present on all the sequenced genomes of each viral group (discontinuous epitopes not included); (b) The epitopes present ≥30 percentage of the sequenced genomes of human seasonal viral groups.
The epitopes borne on the 44 genomes of 2009 H1N1 strains have been examined at the early stage of pandemic outbreak  with the help of this database Immune Epitope Database (IEDB) . No epitopes different from the recent seasonal human influenza virus have been found in the comparison. The subsequent related researches commonly focus on the comparison between novel pandemic strains with seasonal strains [4,17,18], or only on hemagglutinin (HA) protein [19,20] or T-cell epitopes [18,21,22]. The highest proportion of epitopes in the database IEDB is from HA protein, showing that HA protein is the most important B-cell epitope resource in influenza A virus. HA protein contain the regions rich in epitopes that are located at the globular cap of homotrimer and have been divided into four structural sites, Sa and Sb, proximal to the receptor-binding pocket, Ca (Ca1 and Ca2) at the subunit surface and Cb within the vestigial esterase domain . The structural comparison between the 2009 pandemic strain and its closely related 1918 pandemic strain showed that there were 10
Figure 5. Comparison of B-cell epitopes between current seasonal strains H1N1 and H3N2 with the swine corresponding serotypes of strains since 1988. The arrows indicated intertransmission between them.
amino acid alternation in these antigenic regions . We found two epitopes in novel pandemic strains that were not conserved in recent seasonal viruses, one of which was discontinuous epitope in HA protein and located at the antigenic regions Sa and Ca2. However, the six linear epitopes in HA protein were sited outside of those regions, epitope ID 62335 (SVSSFERFEIFPK) (123 - 136, the novel pandemic strain as a reference) close to the region Sa at the cap and others located at stem or close to membrane-based domain (Figure 6). Some epitopes at the cap of HA were not definite in novel pandemic strain but their antibody binding sites have been determined at Sa or close to Sa or Sb regions [14,25]. Therefore, the epitopes at the common antigenic zones of HA globular cap principally did not yet confirmed by experiments in pandemic strain. Recently, a novel epitope between the receptor-binding pocket and the Ca2 antigenic site was found . The epitopes in the globular head typically elicit strain-specific responses because of the hypervariability of this region. It could be predicted that more epitopes will be discovered in the future for this pandemic strain.
Our comparison on NA protein in present study is consistent with that obtained from the research on HA . These authors compared the crystal structures of HA1 and showed that the antigenic region, particularly within the Sa antigenic site, is similar to those of human H1N1 viruses circulating early in the 20th century but much more different from those of recent human seasonal strains. Similarly, our analysis also showed that the epi-
Figure 6. Tridimensional structure of homotrimer HA, indicating the location of common antigenic regions of HA (light blue) and the B-cell epitopes from 2009 pandemic strains (other colors). Only one discontinuous epitope (accession number in IEDB: 91060) (yellow) was located at the common antigenic regions.
topes on NA in 2009 pandemic strains share more similarities to H1N1 strains circulating in 1918-1934 than those in recent seasonal strains. Thus, both proteins (HA, NA) of the most importantly immune effect of the novel H1N1 strains provided the same evidence on human preexisting immunity that can explain why the aged people have less clinical symptom than young ones.
The fact that all the epitopes in the current seasonal human viruses in general population were shared with swine viruses implies that the genetic exchange existed between them. This genetic exchange could be caused by inter-host transmission of strains after chromosomal reassortment or direct inter-host transmission of strains, e.g. the novel pandemic strain. The 2009 pandemic strain H1N1 occurred in April in Mexico and USA in 2009  and sooner it can be found the infection in swine population in Canada and other places of the world later [28- 31]. In 1918, both human and swine populations showed the similar symptoms caused by 1918 pandemic H1N1 virus at similar time so that 1918 pandemic strain resulted from the swine to human or vice verse is still blurred, except the inter-transmission between human and pigs [32-35]. The present epitope analysis showed the coincidence with inter-host genetic exchange, sometimes direct inter-host transmission, of human seasonal influenza viruses with the swine population. However, considering the lack of new human viral epitopes to the general swine virus population (Figure 4(b)), the genetic exchange could occur mostly from swine to human (Figure 5). Multiple reassortment between novel pandemic strains and endemic influenza viruses in pigs has been reported [36,37]. A new swine-origin H3N2 strain that contains M2 gene from 2009 pandemic strain has been recently isolated from several children in Pennsylvania and Indiana . Therefore, such genetic exchange is very active along the evolutionary history.
It is necessary to note that all the data in the IEDB have been confirmed by experiments in the literature. As for the identified epitopes in this study, we only consider the 100% identity of epitopes for comparison. Because of the existence of epitope’s conformational plasticity that allows the alternative amino acids, the actual pre-existing immunity could be higher than our estimate. However, as regard to effectual immunity protection, people still need to face two main problems. One is the new epitopes that could be brought to human by new strains. These new epitopes may come from the genetic drift in animal host by the inter-transmission strains or the gene-reassortant strains. The other is that the virulence related to certain proteins or immunity of epitopes depends on the antibody titers in the infected individuals. As the case of 2009 pandemic influenza, the causing virus contained new epitopes and the infected individuals dominantly produced low titers of the broadly cross-reactive antibodies, which were unable to effectively provide protection from infection. This phenomenon represents a challenge for future research.
One limiting factor for this epitope analysis, based on the comparison of protein amino acid sequences, is not to take into account the glycosylation sites, which could affect the antibody binding or recognition because of the special barrier of glycans. HA and NA both are antigenic glycoproteins and have different glycosylation patterns in different serotypes of strains. Although the potential glycosylation sites could be predicted somehow on the proteins but the actual interaction on the immune recognition needs to be tested by specific experiments.
This work was supported by Secretaría de Investigación y Posgrado del Instituto Politécnico Nacional, México (Grant numbers: 20101372 and 20110896) and by the Consejo Nacional de Ciencia y TecnologíaMéxico (Fondo Mixto de Fomento a la Investigación Científica y Tecnológica, CONACYTGobierno del Estado de Oaxaca. Grant number. 122497). Xianwu Guo, Mario A. Rodríguez-Pérez and Ma. Isabel Salazar hold the scholarships from Comisión de Operación y Fomento de Actividades and Estímulos al Desempeño de los Investigadores/Instituto Politécnico Nacional.