The novel pandemic H1N1 virus has emerged and spread rapidly since early 2009. Although a number of cases of severe pulmonary diseases in children and adults with underlying clinical illness were reported, only mild symptoms are experienced by the majority of people, and the overall case fatality rate is not higher than that of regular seasonal influenza virus infection (5
The observation of a lower infection rate for the 2009 pandemic H1N1 virus in people aged 60 years or older (1
) and the prediction and mapping of cross-reactive CD4 and CD8 T-cell epitopes against this pandemic virus in the general healthy population (12
) prompted us to establish an animal model of preexisting T-cell immunity against 2009 pandemic H1N1 virus infection.
Heterosubtypic T-cell immunity against influenza is a well-established phenomenon in mice. The prototypic priming and challenge models involve the use of the laboratory strain PR8 H1N1 virus and the reassortant X31 virus, which carries the same six internal proteins as PR8 virus and the HA and NA proteins from the 1968 Hong Kong strain (H3N2) of influenza virus (37
). We initially used PR8 H1N1 virus to prime the mice to establish T-cell memory for two major reasons. First, the virus is among the earliest isolated from human influenza virus infections and the protection of aged people from 2009 H1N1 pandemic infection could be due to cross-reactive immunity established after infection with this or a closely related virus circulating in the early 1930s. Second, the PR8 virus has been used widely as a backbone to produce some recombinant human influenza vaccines through reverse genetic techniques, which usually involves the replacement of HA and NA of PR8 virus with those of target viruses that cannot grow well in eggs or cell cultures (41
). Vaccination with these recombinant viruses could thus potentially elicit some level of T-cell response that could cross-react with the 2009 H1N1 virus. Establishing an animal model using the PR8 virus for evaluating cross-protection against 2009 pandemic H1N1 virus would provide direct evidence for the assumptions proposed by earlier studies (12
Previous analysis of human serum samples found neutralizing antibodies against the 2009 pandemic H1N1 virus in people born around 1910, indicating that the 1918 pandemic H1N1 virus is antigenically related to this recent pandemic H1N1 virus (24
). Recently, using an animal model, Garcia-Sastre's lab excellently confirmed that antibodies generated after vaccination with 1918-like virus or 1976 swine H1N1 virus could provide complete protection from lethal infection with 2009 H1N1 virus (27
). Furthermore, cross-reactive monoclonal antibodies against the HA protein of either 1918 virus or CA/09 virus could offer full protection from death by 2009 pandemic virus infection. The importance of eliciting neutralizing antibody for protection from 2009 pandemic H1N1 virus infection has also been indicated in several other animal studies (6
). However, none of these studies carefully examined whether T-cell responses could contribute to the control of 2009 H1N1 influenza virus infection.
Our data reported here also showed that infection with CA/E3/09 virus, but not PR8, X31, or NC/99 virus, could induce potent neutralizing antibody, as demonstrated in vitro
and in vivo
. Our in vitro
microneutralization assay data suggest that neutralizing antibodies are not likely responsible for protection, especially in animals infected with X31 virus, which is a subtype of H3N2. Recent studies indicated that nonneutralizing antibodies, for example, anti-NP antibodies, could also be protective (8
). However, mice that received sera from homosubtypic and heterosubtypic virus-infected mice did not show any signs of protection. Very recently, Suphaphiphat et al. also reported that immune sera collected from ferrets infected with their A/CA/04/09 virus stock had undetectable HAI titers against PR8 virus (36
). In addition, Skountzou et al. reported that sera of mice infected with different H1N1 viruses isolated between 1930 and 1960 had low levels of neutralization activity (HAI titers) against their A/CA/04/09 virus (34
). Thus, our inability to detect cross-reactive HAI activity against CA/E3/09 in PR8-infected sera is essentially comparable with these studies.
It should be pointed out that there are likely some slight variations in antigenicity among the CA/09 stock viruses used in studies from different labs. For example, Skountzou et al. used a CA/09 stock that was mouse adapted after 5 continuous passages, raising the possibility of mutations from the adapting process. Sequencing of our CA/E3/09 virus revealed three mutations in HA (K136N, S200P, and D239G). Due to differences in the numbering system chosen, the S200P mutation corresponds to the S186P mutation in the plasmid F8 genetic mutant CA/09 virus created by Suphaphiphat et al., which has been shown to increase the growth of the virus in either eggs or MDCK cells (36
). The single mutation of S to P at position 186 does not change the antigenicity of the virus (36
). In addition, the K136N mutation in our stock virus is in the same position as the K119N mutation found in an MDCK cell-adapted CA/07/09 virus (11
). The D239G mutation observed in our stock is also the same as the D222G mutation, which results in impaired α2,6-sialic acid (SA) binding and increased recognition of α2,3-SA (11
). To our knowledge, these three mutations occurring together has not been described. Given the low-passage history of our stock, it is likely that this variant may have been present in the seed stock. These results indicate that the HA mutations in our stock contribute to the efficient growth of the virus in eggs and improve its pathogenicity in mice, possibly by altering receptor specificity and SA-binding preference, though antigenicity is preserved, as tested by ferret antisera.
Our CD4 and CD8 T-cell depletion experiments with X31-primed mice indicated that T cells are crucial components of heterosubtypic immunity against the 2009 pandemic influenza virus. Furthermore, CD4 and CD8 T cells contributed equally to protection against 2009 pandemic H1N1 infection. These data are in line with our previous studies and other earlier studies showing that both CD4 and CD8 cells can provide protection against influenza virus infection (7
). Cross-reactive CD8 protection was also illustrated by Skountzou et al. (34
) by the use of priming with their CA/09 virus and rechallenge with A/FM/1/47 or A/Aichi/2/68 virus. We did not perform a T-cell depletion study with PR8-primed mice, but since PR8 and X31 share the same six internal proteins that are major targets of T-cell responses, we believe that the protective effect of CD4 and CD8 cells seen in the X31-primed mice should also be operational in PR8-primed mice after rechallenge.
The protection of NC/99 (H1N1)-primed mice from lethal challenge with CA/E3/09 virus is interesting. The NC/99 virus was continuously used in seasonal influenza vaccines for several years. Recent reports of surveillance data indicated that at least partial levels of protection against the recent pandemic H1N1 virus are associated with prior vaccination with seasonal influenza vaccines (16
). Since in our hands there was no observed neutralization activity or protection by sera from NC/99-primed mice against CA/E3/09 virus, we speculate that the general population having previously received seasonal influenza vaccines containing NC/99 virus could possibly harbor some level of T-cell immunity that might help to reduce disease severity during 2009 pandemic influenza infection.
In summary, our study provides direct evidence that preexisting T-cell immunity could contribute to protection from 2009 pandemic H1N1 infection. Seasonal influenza vaccination that can induce T-cell immunity would be beneficial for current and future pandemic occurrences.