Research ArticleInfluenza

Elicitation of Broadly Neutralizing Influenza Antibodies in Animals with Previous Influenza Exposure

See allHide authors and affiliations

Science Translational Medicine  15 Aug 2012:
Vol. 4, Issue 147, pp. 147ra114
DOI: 10.1126/scitranslmed.3004273


The immune system responds to influenza infection by producing neutralizing antibodies to the viral surface protein, hemagglutinin (HA), which regularly changes its antigenic structure. Antibodies that target the highly conserved stem region of HA neutralize diverse influenza viruses and can be elicited through vaccination in animals and humans. Efforts to develop universal influenza vaccines have focused on strategies to elicit such antibodies; however, the concern has been raised that previous influenza immunity may abrogate the induction of such broadly protective antibodies. We show here that prime-boost immunization can induce broadly neutralizing antibody responses in influenza-immune mice and ferrets that were previously infected or vaccinated. HA stem–directed antibodies were elicited in mice primed with a DNA vaccine and boosted with inactivated vaccine from H1N1 A/New Caledonia/20/1999 (1999 NC) HA regardless of preexposure. Similarly, gene-based vaccination with replication-defective adenovirus 28 (rAd28) and 5 (rAd5) vectors encoding 1999 NC HA elicited stem-directed neutralizing antibodies and conferred protection against unmatched 1934 and 2007 H1N1 virus challenge in influenza-immune ferrets. Indeed, previous exposure to certain strains could enhance immunogenicity: The strongest HA stem–directed immune response was observed in ferrets previously infected with a divergent 1934 H1N1 virus. These findings suggest that broadly neutralizing antibodies against the conserved stem region of HA can be elicited through vaccination despite previous influenza exposure, which supports the feasibility of developing stem-directed universal influenza vaccines for humans.


In a typical year, influenza causes 250,000 hospitalizations and 36,000 deaths in the United States (1). In addition to the morbidity and mortality from seasonal influenza virus, new subtypes may emerge rapidly and unexpectedly, as witnessed with the 2009 H1N1 pandemic outbreak and the threat of the highly pathogenic avian H5N1 virus. These evolving threats underscore the need for more broadly protective influenza vaccines.

Protective immunity against influenza virus in humans arises primarily from neutralizing antibodies directed at the major viral surface protein, hemagglutinin (HA). HA is a trimeric glycoprotein consisting of a highly variable globular head and a conserved stem region. Most monoclonal antibodies (mAbs) elicited by vaccination or infection are directed to the head region of HA, which mediates the attachment of influenza virus to the cellular receptors on the target cells. However, because the globular head of HA undergoes continuous mutations driven by selective pressure, antibodies targeting this region are typically strain-specific and generally do not confer broad protective immunity. In contrast, mAbs directed to the conserved HA stem prevent membrane fusion and broadly neutralize influenza virus from different subtypes. This type of broadly neutralizing antibody (bnAb) was first isolated from mice (2) and recombinant phage display libraries (5-8), and subsequently has been elicited by vaccination in animal models (9) and in humans (10).

Our previous studies have demonstrated the advantage of gene-based vaccination in the induction of cross-protective immunity against influenza virus (9). Priming with an HA DNA vaccine followed by a seasonal vaccine or HA-adenoviral vector boost stimulated the production of antibodies that cross-neutralized various strains of H1N1 virus as well as other influenza A subtypes such as H2N2 and H5N1 in mice, ferrets, and nonhuman primates, and they conferred protection against unmatched viruses in ferrets. H5 HA DNA priming with inactivated vaccine boosting also increased the magnitude of protective antibody responses and, in some cases, induced HA stem–specific antibodies that cross-neutralize other group 1 HAs in humans (10). However, the level and frequency of stem-directed bnAbs circulating in the human sera remain relatively low even after repeated influenza infections and immunizations (11). This finding raised the concern that these bnAbs may be difficult to induce in subjects with previous exposure to influenza. The aim of the current study was therefore to examine whether previous exposure to influenza, by either viral infection or vaccination, alters the induction of stem-directed bnAbs by gene-based prime-boost immunization.


bnAbs elicited by prime-boost immunization in animals with preexisting HA immunity

Mice preinfected with H1N1 A/Puerto Rico/8/1934 (1934 PR8) virus or preimmunized with a replication-defective adenovirus 5 (rAd5) vector expressing 1934 PR8 HA were subjected to gene-based prime-boost immunization with A/New Caledonia/20/1999 (1999 NC) HA (9). We have previously shown that priming with a DNA vaccine encoding 1999 NC HA stimulated the production of stem-directed bnAbs when a matched inactivated vaccine was used as a boost (9). The ability of gene-based prime-boost immunization to induce broader and more potent antibody responses over a traditional inactivated vaccine has been demonstrated in our previous studies in various animal models and in humans (9, 10). Naïve mice that received the same immunization were used as control. A highly sensitive pseudotype HA lentiviral vector system was used to assess neutralizing antibody response because stem-directed antibodies cannot be detected by a hemagglutinin inhibition (HAI) assay, and they are often below the detection limit of a microneutralization assay (9). Prime-boost immunization elicited neutralizing antibodies against both homologous 1999 NC and heterologous HAs, from A/Singapore/6/1986 (1986 SG) and A/Brisbane/59/2007 (2007 Bris), in both naïve and immune mice, whether from infection with H1N1 1934 PR8 virus or from immunization with rAd5–1934 PR8 HA (Fig. 1 and fig. S1A). These data indicated that cross-neutralizing antibodies could be elicited in mice immune to influenza virus by infection or vaccination.

Fig. 1

Neutralizing antibody response in uninfected, previously infected, or immunized mice. Neutralization activity of antisera from uninfected, previously infected, or preimmunized mice was measured using a luciferase reporter with 1999 NC (left), 1986 SG (middle), or 2007 Bris (left) HA-pseudotyped lentiviral vectors. Preexisting HA immunity was established by infection with H1N1 1934 PR8 virus or immunization with rAd5–1934 PR8 HA vaccine as described (see Materials and Methods). The 50% inhibitory concentration (IC50) in neutralization titers (mean ± SD) is shown. Dotted lines indicate the baseline (1:25 dilution) of the pseudotyped lentiviral reporter assay.

To evaluate immunity in a challenge model more relevant to human influenza, ferrets were immunized by infection with H1N1 2007 Bris virus or by immunization with a rAd5 vector expressing 2007 Bris HA. As expected, high titers of neutralizing antibodies were generated against homologous virus after preinfection or preimmunization (fig. S1B, left). Ferrets were also immunized by infection with H1N1 1934 PR8 virus or vaccination with rAd5–1934 PR8 HA. As expected, immune sera from these animals inhibited H1N1 1934 PR8 entry effectively, indicating the presence of neutralizing antibodies to the homologous 1934 PR8 HA (fig. S1B, right). These results demonstrated that preexisting HA immunity was established by either infection or vaccination for each of these viruses in ferrets.

Four weeks after the preinfection or preimmunization, experimental groups were vaccinated with a rAd28 vector followed by a rAd5 vector vaccine encoding 1999 NC HA at weeks 0 and 4, respectively. Antisera were collected 2 weeks after the prime-boost immunization and analyzed for their neutralizing activities. The highest neutralizing titer against 1999 NC HA was observed in the animals preinfected with H1N1 2007 Bris virus, suggesting that preexposure to this virus primed for the 1999 NC HA prime-boost immunization. Animals preimmunized with rAd5–2007 Bris HA vaccine elicited neutralizing antibodies to 1999 NC HA with titers similar to that seen in uninfected animals (Fig. 2A, left). Prime-boost immunization with rAd28/rAd5 HA vaccines also induced cross-reactive neutralizing antibodies in uninfected, H1N1 2007 Bris virus–infected, and rAd5–2007 Bris HA–immunized ferrets; antisera from these groups similarly neutralized an unrelated 1934 PR8 HA pseudotyped virus (Fig. 2B, left). Likewise, preimmunity to HA from an unmatched H1 strain, 1934 PR8, had no effect on the ability of rAd28/rAd5 HA prime-boost immunization to elicit neutralizing antibody responses against homologous 1999 NC HA or heterologous 2007 Bris HA (Fig. 2, A and B, right).

Fig. 2

Neutralizing antibody responses in uninfected, previously infected, or vaccinated ferrets. (A) Preexisting immunity to 2007 Bris HA was established by preinfection of ferrets with 2007 Bris virus or preimmunization with rAd5 vector expressing 2007 Bris HA (left). Similarly, preexisting immunity to 1934 PR8 HA was established by 1934 PR8 virus infection or rAd5 PR8 HA immunization (right). Experimental groups were immunized first with a rAd28 vaccine encoding 1999 NC HA followed by a rAd5 vaccine encoding the same HA with a 4-week interval. Neutralization against 1999 NC HA was measured 2 weeks later. (B) Neutralization of the same antisera against heterologous 1934 PR8 (left) or 2007 Bris (right) HA was measured. Preexisting HA immunity in ferrets was established by infection with 106 EID50 of H1N1 2007 Bris or 1934 PR8 virus or by preimmunization with a rAd5 vector expressing 2007 Bris or 1934 PR8 HA (109 PFU). IC80 titers are shown. IC80 titers >6400 are plotted as 6400. Dotted lines indicate the baseline (1:25 dilution) of the pseudotyped lentiviral reporter assay.

Antibody responses to the globular head of HA were also assessed by HAI in both mice and ferrets. In mice, preimmune sera were negative for HAI against H1N1 1934 PR8, 1999 NC, and 2007 Bris viruses (Table 1, week −4). As expected, mice infected with H1N1 1934 PR8 virus or immunized with rAd5–1934 PR8 HA vector had positive HAI titers to H1N1 1934 PR8 virus (Table 1, week 0, HAI titer ≥1:40) but remained negative for H1N1 1999 NC and 2007 Bris viruses (Table 1, week 0). After immunization with 1999 NC HA, all mice had high titers of HAI antibodies against H1N1 1999 NC and 2007 Bris viruses, indicating that preexisting HA immunity by infection or immunization had no effect on the HAI response induced by prime-boost immunization (Table 1, week 11). The HAI response was also measured in ferret immune sera. In the control groups, HAI titers to H1N1 1999 NC, 1934 PR8, and 2007 Bris viruses were undetectable at baseline (Fig. 3, HAI titer <1:20). In contrast, after rAd28/rAd5 1999 NC HA prime-boost immunization, protective HAI titers against homologous H1N1 1999 NC virus were readily detected in uninfected ferrets and also in animals previously exposed to H1N1 2007 Bris or 1934 PR8 virus or preimmunized with rAd5–2007 Bris or rAd5–1934 PR8 HA vaccines (Fig. 3A, HAI titer ≥1:40). Higher HAI titers against H1N1 1934 PR8 virus developed in animals that had been preexposed to 1934 PR8 HA compared to 2007 Bris (Fig. 3B, right versus left). Similarly, animals previously exposed to 2007 Bris HA by virus infection or immunization generated higher HAI titers against homologous virus compared with 1934 PR8 (Fig. 3C, left versus right). Prime-boost vaccination with 1999 NC HA also elicited low levels of cross-reactive antibodies that recognize the HA head because protective HAI titers against heterologous H1N1 1934 PR8 and 2007 Bris viruses were observed in the uninfected animals (Fig. 3, B and C).

Table 1

HAI antibody titer to H1N1 viruses in uninfected, previously infected, or previously immunized mice that received 1999 NC HA prime-boost immunization.

View this table:
Fig. 3

HAI response in uninfected, previously infected, or immunized ferrets. (A to C) HAI activity of ferret antisera against (A) H1N1 1999 NC, (B) H1N1 1934 PR8, and (C) H1N1 2007 Bris viruses. Dotted lines indicate the minimum protective titer (HAI titer ≥1:40) of the HAI assay.

Elicitation of HA stem–directed antibodies in mice and ferrets

To compare the elicitation of stem-directed antibodies by prime-boost immunization in uninfected animals or those exposed to an unmatched strain by infection or immunization, we analyzed their antisera for the presence of anti-HA stem antibodies. To enrich for this activity by removing those directed to other sites, antisera were first absorbed with 293 cells expressing the stem mutant (Δstem) of 1999 NC HA that blocks binding of CR6261-like stem-specific mAbs (9). Preabsorption removed all non–CR6261-like stem-directed antibodies, and the presence of anti-stem antibodies was then determined by their ability to bind wild-type compared with Δstem 1999 NC HA trimer, as previously described (9). Higher binding to 1999 NC wild-type HA trimer than to the Δstem protein was observed with preabsorbed mouse antisera from prime-boost immunized animals that were either influenza-naïve or preexposed by infection or immunization with H1N1 1934 PR8 virus (Fig. 4A). Similarly, binding of the preabsorbed ferret antisera to wild-type HA trimer was significantly higher than to Δstem HA in all groups that received prime-boost immunization (Fig. 4B), with the strongest anti-stem response in ferrets preinfected with a divergent H1N1 1934 PR8 virus (Fig. 4B, top right). We quantified the level of stem-reactive antibody using mAb CR6261 as a standard and analyzed the relative amounts of stem-reactive antibodies in these ferret immune sera. The HA stem–reactive antibody showed less than a twofold difference in influenza-naïve animals (3.13 to 8.29 μg/ml) compared with those preinfected with 1934 PR8 virus (4.86 to 12.88 μg/ml). Ferrets preinfected with 1934 PR8 virus also generated significantly higher titers of neutralizing antibodies directed to the stem of another subtype virus, H5N1 A/Vietnam/1203/2004 (2004 VN1203) (fig. S2, A and B), thus confirming their breadth, consistent with previous studies (9). These results demonstrated that prime-boost immunization elicited stem-directed antibodies regardless of previous HA exposure.

Fig. 4

Stem-directed antisera elicited by 1999 NC HA prime-boost immunization in mice and ferrets with preexisting HA immunity. (A) Immune sera from uninfected, previously infected, or preimmunized mice were preabsorbed with 293 cells expressing the stem mutant of the 1999 NC HA to remove non–stem-reactive HA antibodies. Binding of preabsorbed sera to wild-type (WT) or stem-mutant (ΔStem) 1999 NC HA was performed by enzyme-linked immunosorbent assay (ELISA). Detection of mouse antibodies was performed with an anti-mouse secondary antibody. Endpoint titers (mean ± SD) are shown. (B) After rAd28/rAd5 1999 NC immunization, ferret antisera from uninfected, previously infected, or preimmunized ferrets were preabsorbed and the binding of preabsorbed sera to WT or stem mutant (ΔStem) 1999 NC HA was performed as described in (A). Detection of ferret antibodies was performed with an anti-ferret secondary antibody. Endpoint titers >6400 are plotted as 6400. Binding of preabsorbed antisera to WT HA from ferrets preinfected with H1N1 1934 PR8 virus (top right, n = 5) versus that from naïve ferrets (bottom left, n = 12). P < 0.0001.

Additional competition assays were performed with mAb CR6261 to document the specificity of these responses. We examined the ability of the CR6261 antibody to block the binding of preabsorbed ferret sera to 1999 NC HA trimer relative to a control immunoglobulin G (IgG). In contrast to an isotype control antibody, CR6261 inhibited the binding of preabsorbed ferret sera to wild-type HA trimers, further confirming the presence of CR6261-like stem-specific antibodies in ferret sera (Fig. 5).

Fig. 5

Specificity of stem-directed antibodies in the ferret antisera. (A to C) ELISA plates coated with WT 1999 NC HA trimers were preincubated with a control IgG or CR6261 mAb before the addition of the preabsorbed ferret antisera described in Fig. 4.

Protection of ferrets against divergent H1N1 viral challenge by prime-boost immunization despite previous HA infection or immunization

We performed the viral challenge study in ferrets because they are generally considered a better animal model to predict vaccine efficacy in humans for influenza (12). Two weeks after the immunization with rAd28/rAd5 vectors encoding 1999 NC HA, ferrets infected with H1N1 2007 Bris virus or immunized with rAd5–2007 Bris HA vector were challenged with heterologous H1N1 1934 PR8 virus (Fig. 6, left). Conversely, ferrets preexposed to H1N1 1934 PR8 virus or rAd5–1934 PR8 HA vaccine were challenged with the unrelated H1N1 2007 Bris virus (Fig. 6, right). Influenza-naïve vaccinated ferrets were challenged with each virus for comparison. Similar to influenza-naïve vaccinated ferrets, prime-boost immunization with 1999 NC HA rAd vectors conferred protection in ferrets that were previously exposed to influenza by either infection or immunization against divergent H1N1 virus, showing a more than two-log reduction in nasopharyngeal viral loads compared to the control animals (Fig. 6).

Fig. 6

Protection of rAd28/rAd5 1999 NC HA immune ferrets from H1N1 1934 PR8 or H1N1 2007 Bris virus challenge despite previous infection or vaccination. Four weeks after the rAd28/rAd5–1999 NC HA prime-boost immunization, ferrets in control and experimental groups were challenged with H1N1 1934 PR8 or 2007 Bris virus (106.5 EID50). Virus titers in the nasal swabs from days 1 and 5 after challenge were determined by means of endpoint titration in Madin-Darby canine kidney cells. Median tissue culture infectious dose (TCID50) (mean ± SD) is shown. Parentheses indicate 1999 NC HA immunization (control naïve versus immune). Dotted lines indicate the minimum detection limit (102 TCID50/ml). * (□) versus (○): P = 0.0048 (left) and P = 0.0007 (right); ** (▪) versus (○): P = 0.0013 (left) and P = 0.0018 (right); *** (♦) versus (○): P = 0.0061 (left) and P = 0.0007 (right).


A variety of mAbs that recognize a highly conserved stem epitope of group 1 HA have been identified in mice and humans (25, 8), providing a rationale for developing more broadly protective influenza vaccines. Antibodies with similarly broad specificity directed toward group 2 influenza virus have also been described, but they bind to a slightly different conserved region on the HA stem (7). The aim of this study was to examine whether such anti-HA stem antibodies can be elicited by vaccination in animals that have been exposed to influenza virus by infection or vaccination where immune memory and patterns of immunodominance may bias the response away from this less immunogenic site.

Recent studies of human sera have suggested that the stem-directed bnAbs can be readily detected in memory B cells or plasma cells but are not readily detected in sera (5, 11). With more potent vaccines, for example, prime-boost combinations or relevant adjuvants, it has been possible to elicit these antibodies in serum by vaccination, and they can confer protective immunity to infectious challenge in mice and ferrets (8, 9, 13). Recently, we have shown that it is possible to elicit such antibodies in humans with H5 HA immunogens, to which most humans have not been exposed and show no previous immunity (10). In contrast, nearly all human adults have been exposed to H1N1 influenza, and it was unknown whether such exposure can affect the induction of stem antibodies in influenza-immune subjects. Our analysis of this question in animals indicates that exposure to influenza, through either infection or vaccination, does not prevent the generation of anti-stem antibodies by gene-based prime-boost immunization. This response in some instances was improved by previous exposure to certain strains: The immune response directed at the conserved stem of influenza HA was greatly increased in ferrets previously infected with H1N1 1934 PR8 virus, although the same effect was not observed with the more closely related H1N1 2007 Bris virus.

We also compared the antibody response in ferrets directed against the head region of influenza HA using HAI assays. As expected, prime-boost vaccination with 1999 NC HA elicits HAI antibodies at a level sufficient to protect against the homologous strain. However, in contrast to broadly neutralizing anti-stem antibodies, previous exposure to heterologous influenza strains did not increase the titer of anti–HA head antibodies except in the case of preexposure to closely related H1N1 2007 Bris. In comparison, previous exposure to a more divergent H1N1 1934 PR8 virus does not improve the HAI response. Together, these results suggest that memory B cells expressing cross-reactive antibodies against the conserved HA stem can be stimulated by more distant viruses, whereas memory B cells expressing antibodies against the antigenically variable HA head region are better stimulated by more closely related virus strains.

Protection by antibodies directed to HA stem in ferrets is likely relevant to influenza immunity in humans. Previous studies have shown that antibodies confer protection against lethal influenza challenge in animals immunized by gene-based vaccination (9, 14, 15). Several groups have also demonstrated protection from lethal H1N1 and H5N1 challenge in mice, using passive transfer of anti–stem HA mAbs such as CR6261, F10, and FE43 (3, 5, 16), thus documenting the ability of such antibodies to confer protection in vivo. Although protection could still possibly be mediated by some other activity of the antibody unrelated to neutralization, there are no data to suggest that the mechanism of stem antibody protection differs from strain-specific, HA head–directed antibodies. Our findings here show that antibodies that neutralize through recognition of the stem region mediate protection, suggesting a similar mode of protection.

We recognize that the considerable sequence and structural variation among different influenza HAs suggest that it is unlikely that prime-boost vaccination with a single HA immunogen could elicit protective immunity against all influenza subtypes. Although most of the anti-stem antibodies against group 1 HAs display considerable neutralization breadth, the reactivity against different subtypes is distinct for different antibodies (5, 8, 16). For instance, mAb FE43 neutralizes H1 subtypes with varying potency (IC50 = 4 to 40 μg/ml) but does not recognize H2 and some H5 subtype viruses; mAb FB110 neutralizes H1, H2, and H5 but does not react with H6 and H9 viruses, whereas another anti-stem mAb, FC41, neutralizes H1, H2, H5, H6, and H9 (5). CR6261 neutralizes H1, H6, H8, and H9 viruses, and it also neutralizes an avian H2 isolate but not the pandemic human H2 virus (16). These observations suggest that subtype or even strain differences within subtype may differentially present the conserved stem epitope and affect neutralization breadth. At the same time, even if the vaccine used in this study were to generate only broader homosubtypic immunity, for example, to H1 subtype viruses, it would add considerable value to the present-day strain-specific vaccines. The next generation of influenza vaccine may be a composite vaccine that provides broad coverage to the current seasonal influenza virus (H1N1, H3N2, and influenza B) as well as other potential pandemic influenza viruses such as H2N2, which has shown its ability to infect humans and cause widespread outbreaks even when it was thought to have disappeared from circulation, analogous to the 2009 H1N1 virus (1, 17).

Our findings suggest that previous exposure alone does not explain why current influenza vaccines fail to generate protective titers of antibodies directed against the conserved stem of influenza HA. The ability of gene-based prime-boost vaccines with novel immunogens and adjuvants to induce neutralizing antibodies that confer more broad influenza immunity in humans will require further clinical evaluation. Nonetheless, our data demonstrate that stem-directed antibodies can be induced in both naïve and HA-preexposed animals, indicating that this vaccine strategy shows promise for the development of a universal influenza vaccine in humans.

Materials and Methods

Plasmid construction

Plasmids encoding different versions of H1N1 and H5N1 HA (1934 PR8, GenBank P03452; 1986 SG, GenBank ABO38395; 1999 NC, GenBank AY289929; 2007 Bris, GenBank ACA28844; 2004 VN1203, GenBank AAT73274) and corresponding neuraminidase proteins and mAb CR6261 were synthesized with human-preferred codons by GeneArt and cloned into a CMV/R expression vector for efficient expression in mammalian cells (18). The stem mutations of 1999 NC HA (Ile545Asn, Gly547Thr) were introduced by site-directed mutagenesis (9).

Production of pseudotyped lentiviral vectors, rAd5, and rAd28

The recombinant lentiviral vectors expressing a luciferase reporter gene were produced as previously described (9). Three replication-defective rAd5 vectors expressing 1999 NC, 1934 PR8, or 2007 Bris HA genes and a replication-defective rAd28 vector expressing 1999 NC HA were produced as previously described (19, 20).

HAI, neutralization, cell absorption, and antibody competition assays

1934 PR8, 1999 NC, and 2007 Bris H1N1 virus seed stocks were expanded in the allantoic cavities of 10-day-old embryonated chicken eggs at 35°C for 48 hours and stored at −80°C. HAI, neutralization, cell absorption, and antibody competition assays were performed as previously described (9). Endpoint titers of antibodies directed against 1999 NC HA were determined by ELISA with recombinant wild-type or Δstem 1999 NC HA trimers using adaptations of previously reported methods (9). mAb CR6261 (4) was produced in 293F cells and purified with a protein G affinity column (GE Healthcare) (9). Endpoint titers were calculated as the most dilute serum concentrations that gave optical density readings of >0.2 above background (9, 10). Titers greater than 6400 were plotted as 6400. CR6261 was used as a standard to measure the level of stem-directed antibodies in the preabsorbed ferret antisera from six naïve animals and five 1934 PR8 virus–preinfected animals. All these animals received prime-boost immunization with 1999 NC HA and were subsequently challenged with 2007 Bris virus. Antibody levels in these preabsorbed sera were calculated at a serum dilution of 1:1600, which falls in the linear range (25 to 75%) of binding to wild-type HA for all samples. Ferret IgG was purified with Nab Protein A/G Spin Kit (Pierce).

Immunizations and challenge

Animal experiments were performed in accordance with all federal regulations and National Institutes of Health (NIH) guidelines. Mouse immunizations with DNA encoding 1999 NC HA and 2006/7 seasonal influenza vaccine (Sanofi Pasteur) were performed as previously described (9). Five mice were used for each group. Before immunization, experimental groups were infected with 0.1 LD50 (median lethal dose) H1N1 1934 PR8 virus or immunized with a rAd5 vector expressing 1934 PR8 HA [108 plaque-forming units (PFU)]. For ferret immunization studies, 6-month-old male Fitch ferrets (Triple F Farms), seronegative for exposure to currently circulating pandemic H1N1, seasonal H1N1, H3N2, and B flu strains, were housed and cared for at BIOQUAL Inc. These facilities are accredited by the American Association for the Accreditation of Laboratory Animal Care International and meet NIH standards as set forth in the Guide for the Care and Use of Laboratory Animals. Forty-eight ferrets were divided into eight groups. Two groups of ferrets were preinfected with 106 egg infective dosage (EID50) of H1N1 2007 Bris (n = 6) or H1N1 1934 PR8 virus (n = 5, one animal was sacrificed because of illness), whereas another two groups of ferrets (n = 6) were preimmunized with rAd5 vectors expressing 2007 Bris or 1934 PR8 HA (109 PFU). All ferrets except the two control groups were immunized with a rAd28 vector expressing 1999 NC HA (109 PFU) at week 0 and boosted with a rAd5 vector expressing the 1999 NC HA (109 PFU) at week 4. The vaccines were administered via intramuscular injections into the upper thigh muscle. Blood was collected 14 days after each immunization and serum was isolated. About 2 weeks after the last immunization, the ferrets were challenged with 106.5 EID50 of H1N1 2007 Bris or H1N1 1934 PR8 virus as previously described (9). Nasal washes were obtained on days 1, 2, 3, 5, 7, 9, and 14 after challenge, and infectious viral titers were determined by a TCID50 by the method of Reed and Muench (21).

Statistical analysis

All data plotted with error bars are expressed as means ± SD. The P values were generated by analyzing data with a two-tailed unpaired t test using the Prism 5 program (GraphPad Software).

Supplementary Materials

Fig. S1. Neutralizing antibody response to homologous HAs in animals previously exposed to influenza by infection or immunization.

Fig. S2. Stem-directed antisera elicited by 1999 NC HA prime-boost immunization in ferrets preinfected with 1934 PR8 virus.

References and Notes

  1. Acknowledgments: We thank S. Rao, A. Taylor, J.-P. Todd, A. Zajac, and C. Chiedi [Vaccine Research Center (VRC)] for help with the animal studies and D. Gordon, A. Perlman, and H. Andersen (BIOQUAL Inc.) for technical support. We thank L. Xu for help with the production of rAd5 HA vectors. We also thank A. Tislerics and B. Hartman for manuscript preparation. Funding: This research was supported by the Intramural Research Program of the VRC, National Institute of Allergy and Infectious Diseases (NIAID), NIH. The findings and conclusions in this report are those of the authors and do not necessarily reflect the views of the funding agency. Author contributions: C.-J.W. and G.J.N. designed the research studies; C.-J.W., H.M.Y., P.M.M., and J.C.B. performed the research; C.-J.W., J.G.D.G., and G.J.N. contributed to development and generation of vectors; C.-J.W., H.M.Y., P.M.M., J.C.B., J.R.R.W., and G.J.N. analyzed the data; and C.-J.W., J.C.B., and G.J.N. wrote the paper. Competing interests: The authors declare that NIH has filed a patent application on work related to gene-based influenza vaccination (DNA prime/inactivated vaccine boost immunization to influenza virus, U.S. Patent 2011/0177122 A1), with C.-J.W. and G.J.N. included as co-inventors. VRC, NIAID, and GenVec Inc. have a cooperative research and development agreement to develop and evaluate adenoviral influenza vaccines. The other authors declare that they have no competing interests.
View Abstract

Navigate This Article