Seroconversion to Seasonal Influenza Viruses after A(H1N1)pdm09 Virus Infection, Quebec, Canada

We looked for cross-reactive antibodies in 122 persons with paired serum samples collected during the 2009 pandemic of influenza virus A(H1N1)pdm09. Eight (12%) of 67 persons with A(H1N1)pdm09 infection confirmed by reverse transcription PCR and/or serology also seroconverted to the seasonal A/Brisbane/59/2007 (H1N1) virus, compared with 1 (2%) of 55 A(H1N1)pdm09-negative persons (p<0.05).

T he role of seasonal 2008-09 trivalent inactivated infl uenza vaccines in protecting against infl uenza A(H1N1)pdm09 virus remains controversial (1). Recent reports indicated that prior infections with seasonal infl uenza A viruses protected against A(H1N1)pdm09 virus infection, suggesting the presence of cross-reactive antibodies (2). Several studies have proposed that humoral immunity and conserved B-and T-cell epitopes contribute to heterosubtypic protection (3,4). Our objective was to determine whether A(H1N1)pdm09 infection induced cross-reactive antibodies against seasonal infl uenza A (H1N1) and A (H3N2) viruses.

The Study
This investigation was part of a trial evaluating A(H1N1)pdm09 transmission among household contacts, conducted during the fi rst wave of the 2009 pandemic (May-July 2009) in Quebec City, Quebec, Canada (5). Clinical data and samples were serially obtained from index case-patients and their contacts in 42 households. Nasopharyngeal secretions were collected from all participants during the fi rst visit and tested by 2 different assays: a conventional reverse transcription PCR (RT-PCR) targeting the hemagglutinin gene of A(H1N1)pdm09 virus (6) and a universal RT-PCR targeting the matrix gene of all infl uenza A viruses (7). Blood was collected from persons >7 years of age at their initial visit (acute-phase sample) and 3-4 weeks later (convalescent-phase sample). Serum was tested by microneutralization assay according to World Health Organization standard protocols with minor modifi cations (8).

Conclusions
During this study, the only infl uenza virus detected in the province of Quebec was A(H1N1)pdm09 virus. Yet, 8 (12%) of 67 A(H1N1)pdm09 virus-infected persons in our study had a concomitant signifi cant increase in microneutralization antibody titers against the most recent A/Brisbane/59/2007(H1N1) strain, of whom 5 persons had 4-8-fold, 2 had 16-fold, and 1 had 32-fold rises. In addition, 4 of these 8 persons also seroconverted to an older A/New Caledonia/20/1999(H1N1) virus, of whom 3 persons had 4-fold and 1 had 16-fold rises between acute-phase and convalescent-phase serum. The cross-reactivity observed in the study population does not seem to be completely subtype specifi c because some persons also showed rising titers against an old infl uenza A (H3N2) strain (A/ Panama/2007/99), although in this case, seroconversion rates did not differ signifi cantly between A(H1N1)pdm09 virus-positive and -negative persons.
A recent study in Hong Kong of 28 paired serum samples showed that infection with the pandemic virus could broaden cross-reactive immunity to other recent subtype H1 swine viruses. In contrast to our study, perhaps because of the small number of participants or older age of A(H1N1) pdm09 virus-positive case-patients (30.5 vs. 20 years), no cross-reactive response was shown against the more recent seasonal infl uenza virus A/HK/400599/2008(H1N1) (9).
We could not determine the extent to which past seasonal infl uenza vaccinations and/or natural infections contributed to the generation of cross-neutralizing antibodies to A(H1N1)pdm09 and the seasonal infl uenza strains. Our next step will be to investigate potential crossneutralizing determinants between these seasonal and pandemic viruses. Neutralizing antibodies that bind to the stalk region of HA2 have been shown to confer broad cross-neutralizing activity against several subtypes of viruses across clades and to provide protection in animal models (10,11). Six of the 8 persons who seroconverted to A/Brisbane/59/2007(H1N1) by microneutralization assay did not meet the 4-fold criteria by hemagglutinin inhibition assay (data not shown), suggesting that the cross-reactivity might result from conserved epitopes in the stalk region of HA2 or in other proteins. Greenbaum et al. recently showed that, overall, 49% of the epitopes reported in recently circulating seasonal A (H1N1) strains were conserved in the A(H1N1)pdm09 virus (12). Specifi cally, 31%, 41%, and 69% of the B-cell, CD4 + and CD8 + T-cell epitopes, respectively, were conserved. Natural infection with A(H1N1)pdm09 virus also could have elicited crossreactive responses against internal components of older viral strains (13). We fi nd intriguing the elaboration of cross-reactive neutralizing antibodies to more recently circulating infl uenza A (H1N1) strains as a result of novel A(H1N1)pdm09 virus infection, whereas the reverse has not generally been evident in serosurveys for cross-reactive A(H1N1)pdm09 antibody, except in elderly persons who had substantial cross-reactive antibodies to A(H1N1) pdm09 virus (14,15). Unfortunately, because of the small sample size of our study and lack of serum from children <7 years of age, we could not assess whether cross-reactivity was an age-dependent phenomenon. However, all but 3 of the cross-reactive seroconverters (13/16 [81%]) were 7-30 years of age. To explore preferential responses to the original infecting virus (original antigenic sin), we assessed crossreactivity for the older A/New Caledonia/99(H1N1) virus that was potentially the priming antigen for some of our younger participants or was closely related to the priming antigen in older participants. However, seroconversion rates for A/New Caledonia (10%) were comparable to those of the more recent A/Brisbane/59/2007(H1N1) strain (12%), and thus we could not distinguish original antigenic sin on that basis. These antigens may have been too closely related antigenically to demonstrate that in this young cohort.
Our work supports the notion that natural A(H1N1) pdm09 virus infection induces broad heterosubtypic (H1 and even H3) responses. It also highlights the need for further investigation of the mechanisms behind crossprotection because they could be keys to creating improved infl uenza vaccines with broader protection.