Sequential Infection in Ferrets with Antigenically Distinct Seasonal H1N1 Influenza Viruses Boosts Hemagglutinin Stalk-Specific Antibodies

Abstract

Broadly reactive antibodies targeting the conserved hemagglutinin (HA) stalk region are elicited following sequential infection or vaccination with influenza viruses belonging to divergent subtypes and/or expressing antigenically distinct HA globular head domains. Here, we demonstrate, through the use of novel chimeric HA proteins and competitive binding assays, that sequential infection of ferrets with antigenically distinct seasonal H1N1 (sH1N1) influenza virus isolates induced an HA stalk-specific antibody response. Additionally, stalk-specific antibody titers were boosted following sequential infection with antigenically distinct sH1N1 isolates in spite of preexisting, cross-reactive, HA-specific antibody titers. Despite a decline in stalk-specific serum antibody titers, sequential sH1N1 influenza virus-infected ferrets were protected from challenge with a novel H1N1 influenza virus (A/California/07/2009), and these ferrets poorly transmitted the virus to naive contacts. Collectively, these findings indicate that HA stalk-specific antibodies are commonly elicited in ferrets following sequential infection with antigenically distinct sH1N1 influenza virus isolates lacking HA receptor-binding site cross-reactivity and can protect ferrets against a pathogenic novel H1N1 virus.

Importance: The influenza virus hemagglutinin (HA) is a major target of the humoral immune response following infection and/or seasonal vaccination. While antibodies targeting the receptor-binding pocket of HA possess strong neutralization capacities, these antibodies are largely strain specific and do not confer protection against antigenic drift variant or novel HA subtype-expressing viruses. In contrast, antibodies targeting the conserved stalk region of HA exhibit broader reactivity among viruses within and among influenza virus subtypes. Here, we show that sequential infection of ferrets with antigenically distinct seasonal H1N1 influenza viruses boosts the antibody responses directed at the HA stalk region. Moreover, ferrets possessing HA stalk-specific antibody were protected against novel H1N1 virus infection and did not transmit the virus to naive contacts.

FIG 1

Sequential infection with antigenically distinct seasonal H1N1 viruses elicits reactivity with novel H1N1 CA/09 in the absence of HAI activity. Groups of ferrets (n = 4) were initially infected with PR/34 and then infected 84 days later with PR/34, Den/57, or Bris/07 sH1N1, as indicated. (A to C) Serum collected at day 14 (D14) and day 84 (D84) following primary (1°) and at day 98 (D98) and day 168 (D168) following secondary (2°) infection was assessed for functional antibodies against the respective H1N1 isolates by HAI. Results are presented as the means ± standard errors of the means of the log₂ HAI titer. Dashed lines denote limits of detection. (D to F) Serum collected at day 14 following primary (1°) and at day 98 following secondary (2°) infection was assessed for binding to CA/09 HA by ELISA. Symbols were used to denote individual ferrets. PR/34, A/Puerto Rico/8/1934; Den/57, A/Denver/1/1957; Bris/07, A/Brisbane/59/2007; CA/09, A/California/07/09.

FIG 2

Antibodies targeting the stalk of HA are boosted following secondary infection with antigenically distinct seasonal H1N1 viruses. (A to C) Serum collected at day 14 and day 84 following primary (1°) and at day 98 and day 168 following secondary (2°) infection was assessed for binding to chimeric HA (cH5/1) expressing the globular head region from H5N1 isolate A/Vietnam/1204/2004 and the stalk regions from CA/09 by ELISA. Results are presented as the means ± standard errors of the means of the log₁₀ endpoint titers (n = 4/time point for all panels). Dashed lines denote limits of detection. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (two-tailed paired t test).

FIG 3

Stalk-specific antibodies are boosted following secondary infection with antigenically distinct seasonal H1N1 viruses. Serum collected at day 14 and day 84 following primary (1°) and at day 98 and day 168 following secondary (2°) infection was assessed for competition with stalk-specific MAb C179 (A to C) or CR6261 (D to F). Results are presented as the means ± standard errors of the means of the log₁₀ 50% inhibition titer (n = 4/time point for all panels). Dashed lines denote limits of detection. *, P < 0.05; **, P < 0.01 (two-tailed paired t test).

FIG 4

Sequential infection with seasonal H1N1 virus boosts stalk specificities despite preexisting antibody cross-reactivity. Groups of ferrets (n = 16) were initially infected with Sing/86 and then infected 84 days later with NC/99 or Bris/07, as indicated. (A and B) Serum collected at day 14 and day 84 following primary (1°) and at day 98 and day 168 following secondary (2°) infection was assessed for functional antibodies against the respective H1N1 isolates by HAI (n = 16/time point). Results are presented as the means ± standard errors of the means of the log₂ HAI titer. Dashed lines denote limits of detection. (C and D) Serum collected at day 14 and day 84 following primary (1°) and at day 98 and day 168 following secondary (2°) infection was assessed for binding to recombinant NC/99 or Bris/07 HA by ELISA (n = 8/time point). (E and F) Serum collected at day 14 and day 84 following primary (1°) and day at 98 and day 168 following secondary (2°) infection was assessed for binding to cH5/1 by ELISA (n = 8/time point). Results in panels C to F are presented as the means ± standard errors of the means of the log₁₀ endpoint titer. Dashed lines denote limits of detection. **, P < 0.01; ***, P < 0.001 (two-tailed paired t test). Sing/86, A/Singapore/6/1986; NC/99, A/New Caledonia/20/1999; Bris/07, A/Brisbane/59/2007; CA/09, A/California/07/2009.

FIG 5

Stalk-specific antibodies are boosted following secondary infection with seasonal H1N1. Serum collected at day 14 and day 84 following primary (1°) and at day 98 and day 168 following secondary (2°) infection was assessed for competition with stalk-specific MAb C179 (A and B) or CR6261 (C and D). Results are presented as the means ± standard errors of the means of the log₁₀ 50% inhibition titer (n = 12 to 14/time point for all panels). Dashed lines denote limits of detection. **, P < 0.01; ***, P < 0.001 (two-tailed paired t test).

FIG 6

Ferrets sequentially infected with historical and modern viruses exhibit a boost in stalk-specific antibody titers following infection with seasonal H1N1 isolates lacking receptor binding site cross-reactivity. Ferrets sequentially infected 12 weeks apart with PR/34 followed by FM/47 and Den/57 or with TX/91 followed by NC/99 and Bris/07 were previously described (22). (A and B) Serum collected at day 14 following primary (1°), at day 98 and day 168 following secondary (2°), and at day 182 and day 252 following tertiary (3°) infection was assessed for binding to chimeric HA (cH5/1) expressing the globular head region from H5N1 isolate A/Vietnam/1204/2004 and the stalk regions from CA/09 by ELISA. Results are presented as the means ± standard errors of the means of the log₁₀ endpoint titers (n = 3/time point for panels A and B). Serum collected at day 14 following primary (1°), at day 98 and day 168 following secondary (2°), and at day 182 and day 252 following tertiary (3°) infection was assessed for competition with stalk-specific MAb C179 or CR6261, as indicated. Results in panels C to F are presented as the means ± standard errors of the means of the log₁₀ 50% inhibition titer (n = 3/time point for panels C to F). Dashed lines denote limits of detection. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (two-tailed paired t test). PR/34, A/Puerto Rico/8/1934; FM/47, A/Fort Monmouth/1/1947; Den/57, A/Denver/1/1957; TX/91, A/Texas/36/1991; NC/99, A/New Caledonia/20/1999; Bris/07, A/Brisbane/59/2007; CA/09, A/California/07/2009.

FIG 7

Reduced viral burden and transmission in ferrets with stalk antibody titers following novel H1N1 virus challenge. (A and B) Ferrets previously infected with seasonal H1N1 (Fig. 1 to 3) were challenged with novel H1N1 (A/California/07/2009). Directly inoculated (D) ferrets were cohoused with an influenza virus-naive contact (C)after 24 h. Viral titers were determined from nasal wash samples collected at day 3 and day 5, as indicated. The solid lines denote geometric mean viral titers for each group. (C and D) Directly inoculated preimmune ferrets were monitored daily for weight loss following infection with a novel H1N1 virus. Results are presented as the means ± standard deviations of individual ferrets within each group (C). Serum collected at day 168 and at day 182 (14 days following CA/09 challenge) from directly inoculated ferrets was assessed for C179 competitive titers (D). Results are presented as the means ± standard errors of the means of the log₁₀ 50% inhibition titer (n = 4/time point). Dashed lines denote limits of detection. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (two-tailed paired or unpaired t test). (E) Serum collected at day 182 from directly inoculated ferrets was assessed for functional antibodies against the respective H1N1 isolates by HAI. Dashed lines denote limits of detection. **, P < 0.01; ***, P < 0.001 (two-tailed unpaired t test).