Swine Flu'' Vaccine Clinical Trials: Cross-reactive Hemagglutinin ...

MAJOR ARTICLE
Revisiting the 1976 ‘‘Swine Flu’’ Vaccine Clinical
Trials: Cross-reactive Hemagglutinin and
Neuraminidase Antibodies and Their Role in
Protection Against the 2009 H1N1 Pandemic
Virus in Mice
Hang Xie, Xing Li, Jin Gao, Zhengshi Lin, Xianghong Jing, Ewan Plant, Olga Zoueva, Maryna C. Eichelberger, and
Zhiping Ye
Laboratory of Respiratory Viral Diseases, Division of Viral Products, Office of Vaccines Research and Review, Center for Biologics Evaluation and
Research, US Food and Drug Administration, Bethesda, Maryland
Background. The 2009 H1N1 pandemic viruses are genetically similar to A/New Jersey/76 H1N1 virus (NJ/76), the
strain selected for the 1976 ‘‘swine flu’’ vaccines. Approximately 45 million people in the United States were vaccinated
against NJ/76 30 years ago, but the impact of this nationwide immunization on the current pandemic is largely unknown.
Methods. Archived human serum samples collected during the 1976 swine flu vaccine trials were assessed for
cross-reactive antibody responses to the 2009 H1N1 pandemic viruses.
Results. Administration of an NJ/76 monovalent vaccine or the combination of a bivalent vaccine (NJ/76 H1N1
and A/Victoria/75 H3N2) plus a B/Hong Kong/72 monovalent vaccine increased hemagglutinin inhibition (HAI) and
neuraminidase inhibition (NAI) antibodies cross-reacting with the 2009 H1N1 pandemic viruses. We showed that
cross-reactive human HAI antibodies elicited by the 1976 swine flu vaccination played a dominant role in protecting
recipient mice against the wild-type A/California/04/2009. Cross-reactive human NAI antibodies were also protective
in recipient mice after a lethal challenge with a hemagglutinin mismatched virus bearing the A/California/04/2009
neuraminidase gene. Transfer of human serum samples with an original HAI titer of 43 or an original NAI titer of 472
was estimated to protect 50% of recipient mice from a lethal infection under the experimental conditions described.
Conclusions. The 1976 swine flu vaccination induced cross-reactive HAI and NAI antibodies that were
functionally protective in mice, suggesting that this vaccination campaign might have had a positive impact on older
adults ($50 years) in the United States during the 2009 H1N1 pandemic.
In 2009, a novel triple reassortant influenza H1N1 virus
of swine origin emerged in Mexico and quickly spread
worldwide, resulting in the first pandemic in 40 years.
Although the majority of the US population was expected
to have little to low preexisting immunity to this
Received 6 April 2011; accepted 26 August 2011; electronically published
5 October 2011.
Correspondence: Hang Xie, PhD, Laboratory of Respiratory Viral Diseases,
Division of Viral Products, Office of Vaccine Research and Review, Center for
Biologics Evaluation and Research, US Food and Drug Administration, 29A Lincoln
Dr, Rm 1B11, Bethesda, MD 20892 (Hang.xie@fda.hhs.gov).
Clinical Infectious Diseases 2011;53(12):1179–87
Published by Oxford University Press on behalf of the Infectious Diseases Society of
America 2011.
1058-4838/2011/5312-0002$14.00
DOI: 10.1093/cid/cir693
novel H1N1 virus [1], the overall mortality associated
with the 2009 influenza H1N1 pandemic was significantly
less than that associated with the 1968 H3N2
‘‘Hong Kong flu’’ and was comparable to that resulting
from seasonal influenza [2]. The cumulative excess
deaths caused by pandemic H1N1 were fewer than expected
among older adults, especially among those aged
$65 years [2]. These observations suggest that older
adults (age, $50 years) might have some crossprotective
immunity from earlier exposure to related
influenza viruses.
Phylogenetic analysis shows that the pandemic H1N1
hemagglutinin (HA) gene is most closely related to that of
A/New Jersey/76-like virus [3], another human virus of
swine origin isolated during a regional outbreak of ‘‘swine
Cross-reactive Hemagglutinin and Neuraminidase Antibodies d CID 2011:53 (15 December) d 1179

flu’’ at Fort Dix military base, New Jersey. The fear that this
outbreak might lead to a pandemic triggered a nationwide vaccination
campaign, resulting in approximately 45 million US
persons being vaccinated against A/New Jersey/76-like virus [4].
However, the impact of this vaccination campaign 30 years ago on
pandemic H1N1 is unclear. We recently retrieved .200 pairs of
human serum samples collected during the 1976 swine flu vaccine
trials from the archives of the Center for Biologics Evaluation and
Research (CBER, formerly the Bureau of Biologics) at the US
Food and Drug Administration (FDA). CBER/FDA, along with
the National Institutes of Health and the Centers for Disease
Control and Prevention, had led the 1976 swine flu vaccine investigations.
Using these archived serum samples, we determined
the cross-reactivity of HA inhibition (HAI) and neuraminidase
(NA) inhibition (NAI) antibodies induced by the 1976 swine flu
vaccines toward pandemic H1N1 viruses and evaluated their
protective potential in naive mice against pandemic H1N1 infection.
METHODS
Ethics Statement
All the human serum samples used in the present study were
analyzed anonymously at CBER/FDA.
Viruses
The following wild-type (wt) viruses were used in the study:
A/New Jersey/8/1976 (H1N1, NJ/76), A/Victoria/3/1975 (H3N2,
Victoria/75), B/Hong Kong/5/1972 (HK/72), A/Brisbane/
59/2007 (H1N1, Brisbane/59), and pandemic H1N1 viruses
A/California/04/2009 (California/04), A/Ontario/RV3226/2009
(Ontario/RV3226), and A/South Carolina/18/2009 (SC/18). The
plasmid expressing the H6 HA gene of A/turkey/Massachusetts/
3740/1965 (H6N2; provided by Dr Doris Bucher, New York
Medical College) was mixed with the pHW2000 plasmids expressing
the NA gene of NJ/76 (N1 NJ) or California/04 (N1 CA)
and the 6 internal genes of A/Puerto Rico/8/34 (H1N1, PR8) to
generate H6N1 NJ or H6N1 CA reassortants by reverse genetics
[5, 6]. The H5 VNN1 CA reassortant was generated also using the
pHW2000 vectors expressing the H5VN HA gene (multibasic
cleavage site deleted) of A/Vietnam/1203/2004 (H5N1, Vietnam/1203),
the N1CA NA gene of California/04, and the 6 internal
genes of PR8. Unless specified otherwise, all wt viruses and
reassortants were amplified in 9-day-old embryonated eggs, and
aliquots were stored at 270°C until use.
Vaccinated Human Serum Samples
Human serum panels (pre- and postvaccination) were collected
during the 1976 swine flu vaccine trials. These clinical trials
involved healthy volunteers aged 18–24 years at the time of the
study. The corresponding clinical protocols were preapproved
1180 d CID 2011:53 (15 December) d Xie et al
by the National Institutes of Health, Bureau of Biologics/FDA,
and George Washington University clinical research review
boards, with written informed consent provided at enrollment.
Unless otherwise indicated, 101 pairs of human serum samples
were retrieved from clinical trials that administered a single dose
of A/New Jersey/76 monovalent vaccine (A/Monovalent) to
participants intramuscularly at 200 chick cell–agglutinating
(CCA) units/dose. Another 120 pairs of human serum samples
were from clinical trials that gave participants an injection (intramuscularly)
of a bivalent A vaccine (A/Bivalent) in the left
arm and a monovalent B vaccine (B/Monovalent) in the right
arm. The A/Bivalent vaccine contained 400 CCA units each of
A/New Jersey/76 (H1N1) and A/Victoria/75 (H3N2)–like viruses
per dose, and the B/Monovalent vaccine contained approximately
500 CCA units/dose of B/Hong Kong/72-like virus. Details
of these clinical trials have been reported elsewhere [7–9].
Hemagglutinin Inhibition Assay
Serum samples were pretreated with receptor-destroying enzyme
(Denka-Seiken) before the HAI assay at a starting dilution
of 1:10 with use of 4 HA units of virus and 0.5% turkey erythrocytes
according to procedures described elsewhere [10, 11].
HAI titers were expressed as the reciprocal of the highest serum
dilution that resulted in complete inhibition of hemagglutination.
A titer of 5 was assigned if no inhibition was observed at
a serum dilution of 1:10.
Neuraminidase Inhibition Assay
A subset of archived human serum samples were randomly selected
from the 1976 A/Monovalent trials (45 pairs) and the
A/Bivalent 1 B/Monovalent combination trials (38 pairs). NAI
titers were determined using an enzyme-linked lectin assay
[12, 13]. In brief, 2-fold serial dilutions of human serum samples
were mixed with an equal volume of a predetermined amount of
H6N1 NJ or H6N1 CA virions in fetuin-coated microtiter plates.
The plates were incubated at 37°C for 16–18 hours After washing,
peroxidase-labeled peanut agglutinin was added, followed by
addition of o-phenylenediamine dihydrochloride 2 hours later.
The color development was stopped by addition of 1 N H 2SO 4
and was read at 490 nm on a Victor V multilabel plate reader
(PerkinElmer). NAI titers were expressed as the reciprocal of the
highest serum dilution that resulted in $50% inhibition of NA
activity. A titer of 5 was assigned if ,50% inhibition of NA
activity was measured at a 1:10 dilution.
Passive Transfer of Human Serum Samples
Ten pairs of human serum samples were selected from the 1976
swine flu vaccine trials for a passive transfer study using the
following criteria: (1) showing at least a 4-fold increase in the
titers of both HAI and NAI antibodies cross-reacting with HA
and NA of California/04 after vaccination and (2) showing no
cross-reactivity toward H5 VN HA of Vietnam/1203 by HAI assay.

Eight-week-old specific-pathogen-free female BALB/c mice
(Taconic Farms) were housed in a biosafety level 2 facility at
CBER/FDA under protocols approved by the Animal Care and
Use Committee of CBER/FDA. Naive mice were ear punched
and were intravenously injected with 100 lL of individual
human sera. Six hours later, recipient mice were challenged
intranasally with two 50% mouse lethal doses (MLD 50) of
Madin-Darby canine kidney–amplified wt California/04 or egggrown
H5 VNN1 CA reassortant. Separate sets of naive mice receiving
100 lL per mouse (intravenously) of pooled California/
04-specific mouse hyperimmune serum samples or Vietnam/
1203-specific mouse hyperimmune serum samples were included
as positive controls. Naive mice receiving 100 lL per
mouse (intravenously) of sterile phosphate-buffered saline
served as negative controls. Body weight and mortality were
monitored daily for 14 days after challenge.
Data Analysis and Statistics
HAI and NAI titers were analyzed for the following: geometric
means of titers for HAI or NAI, seroprotection rates (the proportions
of persons with HAI titers $40), seroconversion rates
(the proportions of persons with a $4-fold increase in HAI
titers), and the proportions of persons with a $4-fold increase in
NAI titers. A correlation analysis between virus-specific HAI
titers and NAI titers was performed using nonparametric
Spearman q test (JMP statistical software, version 7). The correlation
between virus-specific HAI or NAI titers of selected
human serum samples and mouse survival/mortality status after
challenge was analyzed using logistic fit with v 2 .A P value of
,.05 was considered statistically significant.
RESULTS
Cross-reactive Hemagglutinin Inhibition Antibodies Induced by
the 1976 Swine Flu Vaccines
Administration of A/Monovalent or the combination of A/Bivalent
and B/Monovalent resulted in similar NJ/76-specific HAI
geometric mean titers (Table 1), although participants given
A/Monovalent had slightly higher seroprotection rate (23%) and
seroconversion rate (23%) (Figure 1A). In general, the combined
vaccination was not effective at inducing H3 and B strain-specific
HAI antibodies, with only 3% seroprotection and seroconversion
rates against Victoria/75 and HK/72, respectively (Table 1).
Immunization with A/Monovalent vaccine alone or in combination
with H3 and B strains yielded no cross-reactive HAI
antibodies against a recent seasonal H1 vaccine strain, Brisbane/59
(Figure 1B). However, administration of A/Monovalent vaccine
or the combined vaccination induced cross-reactive HAI antibodies
against pandemic H1N1 viruses, including California/04,
Ontario/RV3226, and SC/18 (Figure 1C–E). Participants receiving
the combined vaccination had slightly higher seroprotection and
seroconversion rates for both California/04 and Ontario/RV3226
than did the A/Monovalent recipients (Figure 1C and 1D),
whereas A/Monovalent vaccination resulted in greater
seroprotection and seroconversion rates against SC/18. The
seroprotection and seroconversion rates against SC/18 were
Table 1. Hemagglutination Inhibition Titers Against the Vaccine Strains of 1976 Swine Flu Clinical Trials
Vaccine strain Parameter
NJ/76 Monovalent
(n 5 101)
A/Bivalent 1 B/Monovalent
(n 5 120)
A/New Jersey/8/1976 (H1N1) GMT, prevaccination (95% CI) 5.0 (5.0–5.0) 5.4 (5.1–5.7)
GMT, postvaccination (95% CI) 10.3 (8.2–12.9) 9.5 (7.9–11.6)
% Seroprotection, a prevaccination 0 2
% Seroprotection, postvaccination 23 17
% Seroconversion b
23 15
A/Victoria/3/1975 (H3N2) GMT, prevaccination (95% CI) 5.0 (5–5) 5.1 (5.0–5.1)
GMT, postvaccination (95% CI) 5.2 (4.9–5.4) 6.2 (5.6–6.9)
% Seroprotection, prevaccination 0 0
% Seroprotection, postvaccination 1 3
% Seroconversion 1 3
B/Hong Kong/5/1975 GMT, prevaccination (95% CI) 5.0 (5.0–5.0) 5.0 (5.0–5.1)
GMT, postvaccination (95% CI) 5.0 (5.0–5.0) 5.7 (5.3–6.2)
% Seroprotection, prevaccination 0 0
% Seroprotection, postvaccination 0 3
% Seroconversion 0 3
Abbreviations: CI, confidence interval; GMT: geometric mean titer; HAI, hemagglutinin inhibition.
a The seroprotection rate is the proportion of subjects with HAI titers $40.
b The seroconversion rate is the proportion of subjects with a $4-fold rise in HAI titers.
Cross-reactive Hemagglutinin and Neuraminidase Antibodies d CID 2011:53 (15 December) d 1181

Figure 1. Cross-reactive hemagglutination inhibition (HAI) seroprotection and seroconversion rates after the 1976 swine flu vaccination. Archived
human serum samples collected during the 1976 swine flu vaccination trials were tested by HAI assay using 0.5% turkey erythrocytes. There were 101
pairs of human serum samples from subjects vaccinated with A/New Jersey/76 monovalent vaccine (A/Monovalent) and 120 pairs of human serum
samples from subjects administered with A/New Jersey/76 (NJ/76) and A/Victoria/75 bivalent vaccine (A/Bivalent) in the left arm and B/Hong Kong/75
monovalent vaccine (B/Monovalent) in the right arm (A/Bivalent 1 B/Monovalent). Bars represent virus-specific seroprotection rates (the proportion of
subjects having an HAI titer $40) before and after vaccination. The numbers indicate testing virus-specific seroconversion rates (the proportion of
subjects having a $4-fold increase in HAI titers) for each immunization group. A, HAI response specific for NJ/76; B, HAI response specific for
A/Brisbane/59/2007 (Brisbane/59); C, HAI response specific for A/California/04/2009 (California/04); D, HAI response specific for A/Ontario/RV3226/
2009 (Ontario/RV3226); E, HAI response specific for A/South Carolina/18/2009 (SC/18).
46% and 44%, respectively, for A/Monovalent vaccination
and 26% and 23%, respectively, for the combined vaccination
(Figure 1E).
Cross-reactive Neuraminidase Inhibition Antibodies Induced by
the 1976 Swine Flu Vaccines
Administration of A/Monovalent vaccine alone or in combination
with H3 and B strains significantly elevated NJ/76specific
NAI antibodies from prevaccination levels (P 5 .001 for
A/Monovalent; P , .05 for the combination; Figure 2A). Sixty
percent of the A/Monovalent vaccinees and 32% of the combined
vaccine recipients had a $4-fold increase in NJ/76-specific NAI
titers after vaccination. The A/Monovalent recipients had
significantly elevated California/04-specific cross-reactive NAI
titers, with 44% of them showing a $4-fold increase after
vaccination (P , .001 for A/Monovalent; Figure 2B and data
not shown). In contrast, the combined vaccination did not significantly
increase cross-reactive NAI titers specific for
1182 d CID 2011:53 (15 December) d Xie et al
California/04 despite 33% of the vaccinees having a $4-fold
increase after vaccination (Figure 2B and data not shown). Regardless
of vaccinations, NJ/76-specific or California/04-specific
NAI titers strongly correlated with their corresponding HAI titers
in all enrolled participants (P , .001; data not shown).
Passive Transfer of Cross-reactive Hemagglutinin Inhibition and
Neuraminidase Inhibition Antibodies
Table 2 shows the survival/mortality status of mice that had
received 100 lL per mouse (intravenously) of selected 1976
human serum samples, followed by a lethal challenge with wt
California/04. Of 9 recipient mice that survived the lethal infection
of wt California/04, 7 had received individual human
serum samples with California/04-specific HAI titers $40 and
California/04-specific NAI titers $160 (Table 2). When the
survival/mortality status of all 20 recipient mice in Table 1
was plotted against the original California/04-specific HAI or
NAI titers of each corresponding human serum sample, only

Figure 2. Cross-reactive neuraminidase inhibition (NAI) titers after the 1976 swine flu vaccination. Archived human serum samples collected during the
1976 swine flu vaccination trials were tested for NAI titers using enzyme-linked lectin assay. Forty-five pairs of human sera from subjects vaccinated with
A/New Jersey/76 monovalent vaccine (A/Monovalent) and 38 pairs of human serum samples from subjects administered with A/New Jersey/76 (NJ/76)
and A/Victoria/75 bivalent vaccine plus B/Hong Kong/75 monovalent vaccine (A/Bivalent 1 B/Monovalent) were randomly selected for NAI analysis.
Bars represent virus-specific NAI titers before and after vaccination. A, NAI response specific for NJ/76; B, NAI response specific for A/California/04/
2009 (California/04). *P , .05, **P , .01, ***P , .001 by unpaired Student t test with 2-tailed P value.
Table 2. Survival of Naive Mice That Were Passively Transferred With Human Serum Samples Selected From the 1976 Swine Flu
Vaccine Trials Followed by a Lethal Challenge With Wild-type A/California/04/2009
Serum ID Vaccination
Sera transferred Challenge virus: wt California/04
California/04-specific
HAI titer
California/04-specific
NAI titer
Recipient
mouse ID Survival
17 NJ/76 Monovalent Pre 5 80 1 No
Post 80 320 2 No
39 NJ/76 Monovalent Pre 5 80 3 No
Post 40 640 4 Yes
34 A/Bivalent 1 B/Monovalent Pre 5 160 5 No
Post 40 640 6 Yes
36 A/Bivalent 1 B/Monovalent Pre 5 320 7 No
Post 160 1280 8 Yes
38 A/Bivalent 1 B/Monovalent Pre 10 160 9 No
Post 40 640 10 Yes
52 A/Bivalent 1 B/Monovalent Pre 5 40 11 Yes
Post 80 160 12 Yes
54 A/Bivalent 1 B/Monovalent Pre 5 160 13 No
Post 640 640 14 Yes
55 A/Bivalent 1 B/Monovalent Pre 5 40 15 No
Post 40 160 16 No
63 A/Bivalent 1 B/Monovalent Pre 5 320 17 No
Post 40 1280 18 No
70 A/Bivalent 1 B/Monovalent Pre 5 40 19 Yes
Post 80 160 20 Yes
Pooled California/04-specific mouse hyperimmune sera $320 ND 21–25 Yes (n 5 5)
PBS 26–30 No (n 5 5)
Abbreviations: California/04, A/California/04/2009; HAI, hemagglutination inhibition; ID, identification; NAI, neuraminidase inhibition; ND, not determined; NJ/76,
A/New Jersey/8/1976 H1N1 virus; PBS, phosphate-buffered saline; wt, wild-type.
Naive female Balb/c mice were passively transferred intravenously with 100 lL per mouse of human serum samples selected from the 1976 swine flu vaccine trials.
Six hours later, the mice were challenged intranasally with two 50% mouse lethal doses of wt California/04 (H1N1). Balb/c mice receiving 100 lL per mouse of
California/04-specific mouse hyperimmune sera or 100 lL per mouse of PBS served as positive and negative controls, respectively. Survival was monitored for up to
14 days.
Cross-reactive Hemagglutinin and Neuraminidase Antibodies d CID 2011:53 (15 December) d 1183

Figure 3. Correlation of cross-reactive hemagglutinin inhibition (HAI) and neuraminidase inhibition (NAI) antibodies with protection in recipient
mice against the 2009 H1N1 pandemic viruses after passive transfer of the 1976 swine flu vaccination human serum samples. Ten pairs of archived
human serum samples collected during the 1976 swine flu vaccination trials were selected for passive transfer. Naive female Balb/c mice were
passively transferred (intravenously with 100 lL per mouse of individual human serum sample. Six hours later, the mice were challenged (intranasally
with two 50% mouse lethal doses of wild-type (wt) A/California/04/2009 (California/04) or the H5 VNN1 CA reassortant in A/Puerto Rico/8/34 (PR8)
background bearing the H5 HA gene (multibasic cleavage site deleted) of A/Vietnam/1203/2004 (Vietnam/1203) and the N1 NA gene of California/04.
The correlation between virus-specific HAI or NAI titers of selected human sera and survival/mortality status of recipient mice was analyzed using
logistic fit with v 2 . Blue * indicates survival (S) and red d represents mortality (M) after challenges. A, Correlation between California/04-specific HAI
titer and survival/mortality of recipient mice after challenge with wt California/04; B, Correlation between California/04-specific NAI titer and
survival/mortality of recipient mice after challenge with wt California/04; C, Correlation between Vietnam/1203-specific HAI titer and survival/
mortality of recipient mice after challenge with H5 VNN1 CA reassortant; D, Correlation between California/04-specific NAI titer and survival/mortality
of recipient mice after challenge with H5 VNN1 CA reassortant. The numbers indicate the estimated HAI or NAI titers predictive of 50% survival by
inverse prediction with 1 – a value at .95.
California/04-specific HAI titers, but not NAI titers, correlated
significantly with survival/mortality of these mice (Figure 3A
and 3B). The correlation plot in Figure 3A indicated that transfer
of serum samples with a higher California/04-specific HAI titer
resulted in a greater possibility of survival from a lethal California/04
challenge. According to this correlation model,
transfer of human serum samples with an estimated California/
04-specific HAI titer of 43 (1 2 a 5 .95) predicted 50% survival/
death of recipient mice under the experimental conditions described
(Figure 3A).
To determine whether NA-specific antibodies could contribute
to cross-protection, we evaluated the ability of the archived serum
samples to protect naive mice against H5 VNN1 CA reassortant with
1184 d CID 2011:53 (15 December) d Xie et al
amismatchedHA. The10pairs of selected 1976human serum
samples had no cross-reactive HAI antibodies against H5 VN HA
(HAI titer of 5; Table 3). Six of 7 recipient mice that survived the
lethal H5VNN1CA infection had received human serum samples
with original California/04-specific NAI titers $320 (Table 3).
The survival/mortality status of the H5 VNN1 CA-challenged
mice had no correlation with Vietnam/1203-specific HAI titers
(Figure 3C), whereas a significant correlation was observed
with the original California/04-specific NAI titers of the human
serum samples transferred (Figure 3D). Transfer of human
serum samples with an estimated NAI titer of 472 (1 2 a 5 .95)
predicted 50% survival/death of recipient mice from a lethal
infection of H5 VNN1 CA with a heterologous HA (Figure 3D).

Table 3. Survival of Naive Mice That Were Passively Transferred With Human Serum Samples Selected From the 1976 Swine Flu
Vaccine Trials Followed by a Lethal Challenge With H5VNN1CA Reassortant
Serum ID Vaccination
DISCUSSION
The global control of pandemic H1N1 was likely attributed to
multilevel interventions implemented by individual countries,
including heightened public awareness, application of personal
protective masks and sanitary practices, school closures and social
segregation, enhanced border surveillance, and wide distribution
of effective pandemic vaccines and antivirals [14]. In addition to
these, the 1976 swine flu epidemic might have also contributed to
the reduced severity of pandemic H1N1 in the United States,
because nearly one-quarter of the population was immunized
against NJ/76-like virus [4]. Using archived serum samples from
the 1976 swine flu vaccine trials, we examined the effect of the
1976 swine flu vaccines on cross-reactive HAI and NAI antibodies
and their contributions to protection against pandemic H1N1
infection in mice.
As shown in this study, the 1976 inactivated vaccines, regardless
of formulation, were not very immunogenic, with
Sera transferred Challenge virus: H5VNN1CA
H5VN-specific
HAI titer
N1CA-specific
NAI titer
Recipient
mouse ID Survival
17 NJ/76 Monovalent Pre 5 80 31 No
Post 5 320 32 No
39 NJ/76 Monovalent Pre 5 80 33 No
Post 5 640 34 Yes
34 A/Bivalent 1 B/Monovalent Pre 5 160 35 No
Post 5 640 36 Yes
36 A/Bivalent 1 B/Monovalent Pre 5 320 37 No
Post 5 1280 38 Yes
38 A/Bivalent 1 B/Monovalent Pre 5 160 39 No
Post 5 640 40 No
52 A/Bivalent 1 B/Monovalent Pre 5 40 41 No
Post 5 160 42 No
54 A/Bivalent 1 B/Monovalent Pre 5 160 43 No
Post 5 640 44 Yes
55 A/Bivalent 1 B/Monovalent Pre 5 40 45 No
Post 5 160 46 No
63 A/Bivalent 1 B/Monovalent Pre 5 320 47 Yes
Post 5 1280 48 Yes
70 A/Bivalent 1 B/Monovalent Pre 5 40 49 Yes
Post 5 160 50 No
Pooled Vietnam/1203-specific mouse hyperimmune sera $320 ND 51–55 Yes (n 5 5)
PBS 56–60 No (n 5 5)
Abbreviations: HAI, hemagglutination inhibition; ID, identification; NAI, neuraminidase inhibition; ND, not determined; NJ/76, A/New Jersey/8/1976 H1N1 virus;
PBS, phosphate-buffered saline; Vietnam/1203, A/Vietnam/1203/2004.
Naive female Balb/c mice were passively transferred intravenously with 100 lL per mouse of human serum samples selected from the 1976 swine flu vaccination
trials. Six hours later, the mice were challenged intranasally with two 50% mouse lethal doses of H5VNN1CA reassortant in A/Puerto Rico/8/34 background bearing
the H5VN hemagglutination gene (multibasic cleavage site deleted) of Vietnam/1203 and the N1CA neuraminidase gene of A/California/04/2009. Balb/c mice
receiving 100 lL per mouse of Vietnam/1203-specific mouse hyperimmune sera or 100 lL per mouse of PBS served as positive and negative controls, respectively.
Survival was monitored for up to 14 days.
,30% seroprotection and seroconversion rates for specific
vaccine strains achieved. This is consistent with previous reports
that a single dose of NJ/76 vaccination stimulated a poor HAI
response in participants aged ,25 years [9, 15, 16]. Administration
of NJ/76 monovalent vaccine alone or the combination
of A/Bivalent and B/Monovalent resulted in similar levels of
NJ/76-specific HAI response in all participants, suggesting
that inclusion of H3N2 and B strains had no effect on the
immunogenicity of 1976 swine flu vaccines. Recent clinical
studies have also shown similar immunogenicity of pandemic
A/California/07/2009 (California/07)–containing monovalent
and trivalent vaccines in European adult and older volunteers
[17].
In the present study, we showed that immunization with NJ/
76-like virus resulted in a cross-reactive HAI antibody response
to California/07-like virus (California/04 and California/07
differ only in HA with a D223G change), but not to seasonal
Brisbane/59 strain. This is consistent with the only other report
Cross-reactive Hemagglutinin and Neuraminidase Antibodies d CID 2011:53 (15 December) d 1185

on archived serum samples from persons immunized with
NJ/76 monovalent vaccines [18]. However, no details regarding
participants, vaccines, and immunization were revealed in that
report. Meanwhile, McCullers et al [3] reported that recipients
of the 1976 swine flu vaccines had high cross-reactive but
nonneutralizing antibodies against pandemic H1N1 virus
30 years later. However, all persons enrolled in that study were
health professionals who had been frequently exposed to influenza
viruses or repeated seasonal vaccinations and were
therefore not representative of the general population that received
the 1976 swine flu vaccines. In addition, our results show
that this pandemic H1N1–specific HAI cross-reactivity is not
limited to the California/07-like vaccine strain, but also extends
to its recent variants, including Ontario/RV3226 and SC/18.
Surprisingly, immunization with A/Monovalent vaccine resulted
in a 46% HAI seroprotection rate and 44% seroconversion
rate against SC/18, almost twice the rates specific for the vaccine
strain NJ/76. SC/18 is considered a low reactor with ferret hyperimmune
serum specific for California/07 and is more distant
than Ontario/RV3226 to California/07 in the evolutionary relationships
of HA genes. The observed cross-reactivity of archived
serum samples may also provide an explanation for our previous
finding that 10%–15% of participants ($50 years) enrolled in
the US 2009–2010 seasonal clinical trial had a baseline HAI titer
of $40 against SC/18 [19].
In addition to cross-reactive HAI response, we also examined
the presence of cross-reactive NAI antibodies in these archived
serum samples. It is well documented that NAI antibodies can be
induced by natural infections or immunization and play a role in
protection against influenza infection [20–26]. The 1976 swine
flu vaccines contained NA at various levels [24], and the NA
genes of NJ/76 and California/04-like viruses are 82% identical in
their amino acid sequences. Thus, it is not surprising that immunization
against NJ/76 resulted in enhanced NAI responses
that cross-reacted with pandemic H1N1 virus. The augmentation
of virus-specific NAI response after A/Monovalent vaccination
was higher than that after administration of the combined vaccines,
implying that NA from additional strains might not synergistically
enhance the immunogenicity of NA.
NA-specific antibodies can block the release of newly synthesized
viral particles, reduce plaque sizes, and decrease virus
yields in vitro but are unable to prevent infection in vivo
[20, 27]. Previous studies have shown that NAI antibodies could
inhibit pulmonary virus replication, reduce lung lesions in infected
mice in the absence of a matched HAI response, and ease
illness in human volunteers challenged with viruses bearing
antigenically identical NA genes [21, 23, 28]. Murine hyperimmune
serum samples derived from DNA vaccination against
seasonal N1 NA provided some protection in recipient mice
against H5N1 challenge [25], suggesting that cross-reactive
NAI antibodies might potentially enhance a host’s resistance to
1186 d CID 2011:53 (15 December) d Xie et al
heterologous influenza strains even without cross-reactive HAI
responses. Marcelin et al [26] reported that seasonal vaccination
enhanced pandemic H1N1–specific cross-reactive NAI antibodies
in US adults and older persons, predominantly in those
aged $60 years, which might have potentially contributed to the
reduced severity of the 2009 pandemic in older adults.
In the current study, we directly addressed the role of crossreactive
human HAI and NAI antibodies elicited by the 1976
swine flu vaccines in protection against pandemic H1N1 infection
in mice. Our data clearly indicate that protection
against wt California/04 was HAI dependent. We also directly
demonstrate that human HAI antibodies elicited by the 1976
swine flu vaccination cross-protects against pandemic H1N1 virus
in a mammalian species. In our model, transfer of human
serum samples with an estimated HAI titer of 43 was predicted to
provide 50% protection in recipient mice. An HAI titer of 40 is an
established seroprotective measure that is considered to be associated
with at least 50% reduction in the risk of influenza
infection or disease in humans [29, 30]. The consistency between
our estimate and the well-accepted seroprotective measure validates
our model. Although there was no correlation of NAI titers
with protection against wt California/04 in the presence of
a cross-reactive HAI response, the NAI titers strongly correlated
with survival of recipient mice challenged with a virus containing
California/04 NA gene but a mismatched HA gene. In addition,
we found that these selected 1976 human serum samples contained
NAI antibodies cross-reacting with N3 NA and protected
recipient mice from the lethality of an H5 VNN3 reassortant (data
not shown). The 9 NA subtypes of influenza A fall into 2 distinct
groups in the phylogenetic tree, with N1 in group 1 and N2 and
N3 in group 2 separately [31]. The aforementioned crossreactivity
is not surprising, because the 1976 swine flu vaccine
recipients had been exposed to either an H3N2 virus or an H3N2containing
vaccine. Currently, there is no regulation of NA
content in US licensed seasonal influenza vaccines. Our present
study shows a potential benefit of cross-reactive NAI antibodies
when a matched HAI response is deficient, thus urging further
investigations in this topic and consideration of NA standardization
in new influenza vaccines.
Taken together, our study demonstrates that the 1976 swine
flu vaccination induced cross-reactive HAI and NAI antibodies
that were functionally protective in mice, which might have had
a positive effect on older adults in the United States during the
influenza H1N1 pandemic.
Notes
Acknowledgments. We thank Dr Xiyan Xu for providing seasonal and
pandemic influenza viruses; Dr Doris Bucher for providing the plasmid
expressing the H6 HA gene of A/turkey/Massachusetts/3740/1965;
Dr Matthew Sandbulte for generation of the H6N1CA reassortant; and
Drs Carol Weiss and Marian Major for critical review of this manuscript.

The findings and conclusions in this article have not been formally
disseminated by the Food and Drug Administration and should not be
construed to represent any Agency determination or policy.
Financial support. This work was supported by the institutional research
funds provided by CBER/FDA.
Potential conflicts of interest. All authors: No reported conflicts.
All authors have submitted the ICMJE Form for Disclosure of Potential
Conflicts of Interest. Conflicts that the editors consider relevant to the
content of the manuscript have been disclosed.
Author Contributions. Conceived and designed the experiments: H. X.
Provided the 1976 human sera: Z. Y. Performed the HAI assays: X. X., X. L.,
Z. L., X. J., E. P., O. Z., Z. Y. Generated the H6N1NJ reassortant for NAI
assays: J. G., M. E. Performed the NAI assays: J. G., M. E. Generated the
challenge viruses: H. X., X. L. Conducted the animal experiments: H. X.,
Z. L. Analyzed the data: H. X. Performed the statistical analysis: H. X.
Wrote the paper: H. X.
References
1. Centers for Disease Control and Prevention. Serum cross-reactive antibody
response to a novel influenza A (H1N1) virus after vaccination with
seasonal influenza vaccine. MMWR Morb Mortal Wkly Rep 2009;
58:521–4.
2. Armstrong GL, Brammer L, Finelli L. Timely assessment of the severity
of the 2009 H1N1 influenza pandemic. Clin Infect Dis 2011; 52(Suppl 1):
S83–9.
3. McCullers JA, Van De Velde LA, Allison KJ, Branum KC, Webby RJ,
Flynn PM. Recipients of vaccine against the 1976 ‘‘swine flu’’ have
enhanced neutralization responses to the 2009 novel H1N1 influenza
virus. Clin Infect Dis 2010; 50:1487–92.
4. Sencer DJ, Milar JD. Reflections on the 1976 swine flu vaccination
program. Emerg Infect Dis 2006; 12:29–33.
5. Hoffmann E, Neumann G, Kawaoka Y, Hobom G, Webster RG.
A DNA transfection system for generation of influenza A virus from
eight plasmids. Proc Natl Acad Sci U S A 2000; 97:6108–13.
6. Sandbulte MR, Gao J, Straight TM, Eichelberger MC. A miniaturized
assay for influenza neuraminidase-inhibiting antibodies utilizing reverse
genetics-derived antigens. Influenza Other Respi Viruses 2009; 3:233–40.
7. Dolin R, Wise TG, Mazur MH, Tuazon CU, Ennis FA. Immunogenicity
and reactogenicity of influenza A/New Jersey/76 virus vaccines in
normal adults. J Infect Dis 1977; 136(Suppl):S435–42.
8. Wise TG, Dolin R, Mazur MH, Ennis FA. Serologic responses after two
sequential doses of influenza A/New Jersey/76 virus vaccine in normal
young adults. J Infect Dis 1977; 136(Suppl):S496–9.
9. Wise TG, Dolin R, Mazur MH, Top FH Jr, Edelman R, Ennis FA.
Serologic responses and systemic reactions in adults after vaccination
with bivalent A/Victoria/75-A/New Jersey/76 and monovalent B/
Hong Kong/72 influenza vaccines. J Infect Dis 1977; 136(Suppl):
S507–17.
10. Kendal AP, Pereira MS, Skehel JJ. Hemagglutination inhibition. In:
Kendal AP, Pereira MS, Skehel JJ, eds. Concepts and procedures for
laboratory-based influenza surveillance. Atlanta, GA: Centers for Disease
Control and Prevention, 1982; B17–35.
11. Rowe T, Abernathy RA, Hu-Primmer J, et al. Detection of antibody to
avian influenza A (H5N1) virus in human serum by using a combination
of serologic assays. J Clin Microbiol 1999; 37:937–43.
12. Lambre CR, Terzidis H, Greffard A, Webster RG. Measurement of antiinfluenza
neuraminidase antibody using a peroxidase-linked lectin and
microtitre plates coated with natural substrates. J Immunol Methods
1990; 135:49–57.
13. Cate TR, Rayford Y, Nino D, et al. A high dosage influenza vaccine
induced significantly more neuraminidase antibody than standard
vaccine among elderly subjects. Vaccine 2010; 28:2076–9.
14. Leung GM, Nicoll A. Reflections on pandemic (H1N1) 2009 and the
international response. PLoS Med 2010; 7:e1000346.
15. Parkman PD, Hopps HE, Rastogi SC, Meyer HM Jr. Summary of
clinical trials of influenza virus vaccines in adults. J Infect Dis 1977;
136(Suppl):S722–30.
16. Foy HM, Cooney MK, Taylor J, et al. Antibody response to influenza
A/New Jersey and A/Victoria virus vaccines in 1976 and subsequent
antibody levels after influenza A epidemics, 1977–1979. J Infect Dis
1980; 142:139–44.
17. Van der Vliet D, Pepin S, Lambert M, et al. Similar immunogenicity
and safety of the A/H1N1 2009 pandemic influenza strain when used as
a monovalent or a trivalent vaccine. Hum Vaccin 2010; 6:823–8.
18. Hancock K, Veguilla V, Lu X, et al. Cross-reactive antibody responses to the
2009 pandemic H1N1 influenza virus. N Engl J Med 2009; 361:1945–52.
19. Xie H, Jing X, Li X, et al. Immunogenicity and cross-reactivity of 2009–
2010 inactivated seasonal influenza vaccine in US adults and elderly.
PLoS One 2011; 6:e16650.
20. Webster RG, Laver WG. Preparation and properties of antibody directed
specifically against the neuraminidase of influenza virus.
J lmmunol 1967; 99:49–55.
21. Schulman JL, Khakpour M, Kilbourne ED. Protective effects of specific
immunity to viral neuraminidase on influenza virus infection of mice.
J Virol 1968; 2:778–86.
22. Monto AS, Kendal AP. Effect of neuraminidase antibody on Hong
Kong influenza. Lancet 1973; 1:623–5.
23. Couch RB, Kasel JA, Gerin JL, Schulman JL, Kilbourne ED. Induction
of partial immunity to influenza by a neuraminidase-specific vaccine.
J Infect Dis 1974; 129:411–20.
24. Kendal AP, Noble GR, Dowdle WR. Neuraminidase content of influenza
vaccines and neuraminidase antibody responses after vaccination
of immunologically primed and unprimed populations. J Infect
Dis 1977; 136(Suppl):S415–24.
25. Sandbulte MR, Jimenez GS, Boon AC, Smith LR, Treanor JJ, Webby RJ.
Cross-reactive neuraminidase antibodies afford partial protection
against H5N1 in mice and are present in unexposed humans. PLoS
Med 2007; 4:e59.
26. Marcelin G, Bland HM, Negovetich NJ, et al. Inactivated seasonal influenza
vaccines increase serum antibodies to the neuraminidase of
pandemic influenza A(H1N1) 2009 virus in an age-dependent manner.
J Infect Dis 2010; 202:1634–8.
27. Kilbourne ED, Laver WG, Schulman JL, Webster RG. Antiviral activity
of antiserum specific for an influenza virus neuraminidase. J Virol
1968; 2:281–8.
28. Murphy BR, Kasel JA, Chanock RM. Association of serum antineuraminidase
antibody with resistance to influenza in man. N Engl
J Med 1972; 286:1329–32.
29. Potter CW, Oxford JS. Determinants of immunity to influenza infection
in man. Br Med Bull 1979; 35:69–75.
30. Fox JP, Hall CE, Cooney MK, Foy HM. Influenza virus infections in
Seattle families, 1975–1979. I. Study design, methods and the occurrence
of infections by time and age. Am J Epidemiol 1982; 116:212–27.
31. Gamblin SJ, Skehel JJ. Influenza hemagglutinin and neuraminidase
membrane glycoproteins. J Biol Chem 2010; 285:28403–9.
Cross-reactive Hemagglutinin and Neuraminidase Antibodies d CID 2011:53 (15 December) d 1187