Recombinant influenza H7 hemagglutinins induce lower neutralizing antibody titers in mice than do seasonal hemagglutinins

Abstract

Background: Vaccines against avian influenza viruses often require high hemagglutinin (HA) doses or adjuvants to achieve serological titers associated with protection against disease. In particular, viruses of the H7 subtype frequently do not induce strong antibody responses following immunization.

Objectives: To evaluate whether poor immunogenicity of H7 viruses is an intrinsic property of the H7 hemagglutinin.

Methods: We compared the immunogenicity, in naïve mice, of purified recombinant HA from two H7 viruses [A/Netherlands/219/2003(H7N7) and A/New York/107/2003(H7N2)] to that of HA from human pandemic [A/California/07/2009(H1N1pdm09)] and seasonal [A/Perth16/2009(H3N2)] viruses.

Results: After two intramuscular injections with purified hemagglutinin, mice produced antibodies to all HAs, but the response to the human virus HAs was greater than to H7 HAs. The difference was relatively minor when measured by ELISA, greater when measured by hemagglutination inhibition assays, and more marked still by microneutralization assays. H7 HAs induced little or no neutralizing antibody response in mice at either dose tested. Antibodies induced by H7 were of significantly lower avidity than for H3 or H1N1pdm09.

Conclusions: We conclude that H7 HAs may be intrinsically less immunogenic than HA from seasonal human influenza viruses.

Keywords: H7 influenza viruses; hemagglutinin; immunogenicity; influenza; influenza pandemics; influenza vaccines.

Figure 1

ELISA responses to recombinant hemagglutinins. Mice were immunized with (A, B) 3 μg or (C, D) 10 μg of purified recombinant hemagglutinin (HA) from Perth/16, CA/07, NL/219, or NY/107 on two occasions 21 days apart. Sera collected on days 21, 35, 49, and 63 after the initial immunization were tested by ELISA, using (A, C, E) purified recombinant homologous full-length HA or (B, D, F) HA1 as antigen. (E, F) Same experiment as A–D, showing the geometric mean titer (GMT) of the highest titer achieved by each mouse at any time point. Data are shown as GMT for each group of mice. Error bars represent 95% confidence intervals. Statistically significant differences are indicated by * (0·0001 < P < 0·05) or ** (P < 0·0001). Data are from one representative experiment of 3 replicates.

Figure 2

Hemagglutination inhibition (HI)responses to recombinant hemagglutinin. Mice were immunized with (A) 3 μg or (B) 10 μg of purified recombinant hemagglutinin from Perth/16, CA/07, NL/219, or NY/107 on two occasions 21 days apart. Sera collected on days 21, 35, 49, and 63 after the initial immunization were tested by HI assays (limit of detection: titer of 10), using homologous live virus as antigen. (C) Same experiment as A and B, showing the geometric mean titer (GMT) of the highest titer achieved by each mouse at any time point. Values are shown as GMT. Error bars represent 95% confidence intervals. Differences between groups are indicated by * (0·0001 < P < 0·05) or ** (P < 0·0001). Data are from one representative experiment of 3 replicates

Figure 3

Microneutralization responses to recombinant hemagglutinin. Mice were immunized with (A) 3 μg or (B) 10 μg of purified recombinant hemagglutinin from Perth/16, CA/07, NL/219, or NY/107 on two occasions 21 days apart. On days 21, 35, 49, and 63 after the initial immunization, serum antibody titer was tested by microneutralization (MN) assays, using homologous live virus as antigen (limit of detection: titer of 20). (C) Same experiment as A and B, showing the geometric mean titer (GMT) of the highest titer achieved by each mouse at any time point. (D) MN and HI values for positive-control ferret sera, from ferrets infected with live NL/219 or NY/107. Values are shown as GMT. Error bars represent 95% confidence intervals. Differences between groups are indicated by * (0·0001 < P < 0·05) or ** (P < 0·0001). Data are from one representative experiment of 3 replicates.

Figure 4

Hemagglutination inhibition (HI) and microneutralization (MN) titers and avidity ELISA for mice with HI titers between 80 and 160. Serum samples with HI titers between 80 and 160 were selected (n = 6 mice per group). (A) geometric mean titer (GMT) HI and MN titers for each group. (B) Avidity ELISAs were performed with the addition of 5-min incubation with either PBST or 4M urea. Values are shown as GMT. Differences between PBS and urea-treated samples are indicated by ‘†’ (P < 0·05). (C) The avidity index (AI) was calculated for each sample. Values are shown as average AI (percent). Error bars represent 95% confidence intervals. Differences between groups are indicated by * (0·0001 < P < 0·05) or ** (P < 0·0001).