Vaccine 28 (2010) 523–531
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/vaccine
Swine influenza matrix 2 (M2) protein contributes to protection against
infection with different H1 swine influenza virus (SIV) isolates
Pravina Kitikoona, Amy L. Vincentb, Bruce H. Jankea, B. Ericksona, Erin L. Straita,
Shan Yua, Marie R. Gramerc, Eileen L. Thackera,∗
aCollege of Veterinary Medicine, Iowa State University, Ames, IA, USA
bVirus and Prion Diseases of Livestock Research Unit, National Animal Disease Center, ARS-USDA, Ames, IA, USA
cMinnesota Veterinary Diagnostic Laboratory, University of Minnesota, Saint Paul, MN, USA
a r t i c l ei n f o
Received 12 March 2009
Received in revised form
18 September 2009
Accepted 30 September 2009
Available online 3 November 2009
Swine influenza virus
a b s t r a c t
A swine influenza virus (SIV) vaccine-challenge pig model was used to study the potential of a conserved
matrix 2 (M2) protein vaccine alone or in combination with an inactivated H1N1-vaccine to protect
against H1N1 and H1N2 viruses. The H1N1-vaccine and heterologous H1N2-challenge virus model has
previously been shown to prolong fever and increase SIV-associated pneumonic lesions. The M2 vaccine
in combination with the H1N1-vaccine reduced the H1N2 induced fever but not virus shedding. The
M2 vaccine alone reduced respiratory signs and pneumonic lesions to levels similar to the negative
control pigs following H1N2 infection. This study found that the M2 protein has potential as a vaccine
for SIV-associated disease prevention. However, development of an immune response towards the major
envelope HA protein was required to reduce SIV shedding.
© 2009 Elsevier Ltd. All rights reserved.
Influenza A virus causes respiratory disease in avian and mam-
malian species . Immunization with inactivated influenza virus
vaccines is currently the most common method used to prevent
influenza-associated disease in the human, equine and swine pop-
ulations. Protection induced by these vaccines has been associated
with the induction of antibodies to the two major viral envelope
tle to no heterologous cross-reactive immunity (Het-I) between
influenza subtypes or genetically diverse viruses within subtype
is induced by these vaccines . As a result, influenza A vaccines
are often manufactured as a bi- or trivalent vaccines containing
genetically different types and/or subtypes of the viruses circulat-
ing in each host population. Vaccine efficacy is directly correlated
with how closely the vaccine virus matches the field viruses within
each subtype [3,4]. Mismatch of the vaccine and field viruses is a
frequent cause for vaccine failure in the field. Because influenza
type A viruses undergo antigenic drift, human influenza vaccines
are re-evaluated yearly. The viruses used in human vaccines are
determined by intensive field surveys with laboratory evaluations
involving the collaboration of many authorities and organizations
around the world.
∗Corresponding author. Tel.: +1 301 504 5774.
E-mail address: email@example.com (E.L. Thacker).
The emergence of a new swine influenza viruses (SIV) in the
United States swine population in the late 1990s has complicated
the development of efficacious vaccines for pigs . The entrance
of the H3N2 virus into the US swine population has resulted in
the generation of new genetically diverse viruses due to viral
reassortment. The presence of multiple genetically diverse viruses
within each subtype has reduced the success of vaccine efficacy
in swine. The importance of matching virus strains within sub-
types was recently demonstrated in Europe where a study found
that a bivalent H1N1/H3N2 vaccine provided suboptimal protec-
tion in pigs against a H1N2 challenge . Interestingly, a separate
study observed disease enhancement in pigs that were vaccinated
with an inactivated H1N1 virus followed by infection with a genet-
ically different H1N2 virus . Unlike human influenza vaccines,
SIV vaccines are not revised annually. Maintaining up to date SIV
vaccines would require the same active surveillance and networks
of laboratories requiring extensive funding similar to required for
human vaccines. Similar laboratory networks and funding is not
currently available to the swine industry, reducing the response
time to newly emerging viruses.
The concept of a “universal influenza vaccine” based on the con-
served matrix 2 (M2) minor envelope protein of the influenza A
virus has been a focal point for influenza-associated-disease pre-
vention research [8,9]. The M2 protein forms a proton channel,
which is essential in uncoating the virus during the initial stage
of infection [10,11]. It is a transmembrane protein composed of a
non-glycosylated ectodomain (M2e) made up of 24 amino acids at
0264-410X/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
P. Kitikoon et al. / Vaccine 28 (2010) 523–531
the N-terminus, 19 amino acids spanning the lipid bilayer and 54
amino acids make up the C-terminus cytoplasmic tail . Studies
with influenza A and B viruses have shown that antibodies against
the extracellular domain of the M2 protein inhibit virus replication
in MDCK cells  and M2e-specific antibodies decrease the viral
load in mouse lungs . Passively transferred antibodies to the
strated 90–100% protection against lethal challenge with influenza
virus in mice .
M2 vaccine-challenge studies conducted in animal models that
are not natural hosts for influenza A infection were promising.
Several studies in mice found that M2e is capable of inducing anti-
bodies that reduce clinical disease and prevent lethal challenge
[14,16,17]. Another vaccine study using a M2 peptide conjugate
conducted in mice, ferrets and rhesus monkeys found that the vac-
cinated animals shed reduced levels of virus in nasal secretions
and had reduced amounts of viral antigen in the lungs . An
early study in pigs, a natural host of influenza A viruses, suggested
that M2e may play a role in heterosubtypic immunity as low levels
of M2e-antibodies were detected in pig’s sera following a primary
H3N2 influenza infection and much higher levels were observed
subsequently with infection with a H1N1 virus . In contrast,
one experiment investigated if the induction of the cell mediated
immune (CMI) response based on two conserved influenza A pro-
teins, M2 and nucleoprotein (NP) provided protection against SIV
infection . This study found that pigs vaccinated with a DNA-
construct expressing the M2e–NP fusion protein and infected with
SIV lacked antibodies to the HA and NA protein and had enhanced
disease. These results suggest that the presence of a CMI induced
response to the M2e and NP proteins alone without the presence
of antibodies against the two major envelope glycoproteins (HA
and NA proteins) is insufficient to provide protection against SIV
infection and may even result in a negative outcome.
The purpose of this study was to investigate the efficacy of a
M2 protein vaccine in combination with an inactivated H1N1 virus
in preventing SIV vaccine-associated disease enhancement in pigs
using a previously described vaccine-heterologous virus challenge
model. This model demonstrated SIV vaccine-associated disease
enhancement in an earlier study . This model was used in this
study as SIV vaccine-induced disease enhancement can potentially
occur in the field from mismatched vaccine and field viruses. Our
primary objective was to study the efficacy of the M2 protein when
used in combination with a monovalent H1N1 inactivated vaccine
in protecting against a heterologous subtype (H1N2) SIV infection
compared to the protection induced by the M2 protein alone.
2. Materials and methods
2.1. Experimental design
2.1.1. Virus and vaccine preparation
The virus used to prepare the inactivated SIV vaccine was
A/Sw/IA/15/1930 H1N1 (IA30). Viruses used as challenge inoc-
ula were the homologous virus (IA30) and a heterologous virus,
A/Sw/MN/00194/2003 H1N2 (MN03). Both viruses had been used
in a previously described experimental vaccine-challenge model
. All viruses were propagated in Madin-Darby canine kidney
(MDCK) cells. The virus was harvested and cell debris clarified by
centrifugation prior to inoculating pigs. Pigs were inoculated intra-
tracheally with a dose of 5ml of 1×10750% tissue culture infective
dose (TCID50)/milliliter (ml).
The IA30 inactivated vaccine was prepared as described previ-
ously . Briefly, IA30 virus with a hemagglutinin unit of 400 per
cells. A commercial adjuvant was added to the inactivated virus at
a ratio of 1:1 (Emulsigen, MVP Laboratories, Inc., Ralston, NE).
The recombinant M2 (rM2) protein was produced using a bac-
ulovirus expression system and isolated as described previously
. A concentration of 50?g/ml of rM2 protein was combined
with the same adjuvant used with the IA30 inactivated vaccine.
All study procedures and animal care activities were conducted
in accordance with the guidelines and under the approval of the
Forty-eight twelve-day-old pigs obtained from a conventional
herd serologically free of SIV, porcine reproductive and respira-
allotted randomly to 9 treatment groups. The experimental design
is described in Table 1. Throughout the study, pigs were housed in
identical isolation rooms based on their challenge status. Pigs were
provided feed and water ad libitum.
Pigs in the appropriate groups received an inactivated SIV vac-
cine described earlier and/or a recombinant M2 protein vaccine at
vaccine were administered intramuscularly (IM) to pigs in groups
2–4 and 6–8 at a dose of 2ml per pigs. Groups assigned to receive
both the inactivated SIV vaccine and rM2 vaccine were injected IM
at different sites. Pigs were inoculated intratracheally with either
the homologous virus (vaccine strain) or a heterologous virus at 7
weeks of age (Trial day 0).
Nasal swabs were collected at −1, 2, 3, and 5 days post infection
(DPI). Blood was collected at −29, −22, −15, −8, −1 and 5 DPI. Sera
was stored at −20◦C and assayed simultaneously at the end of the
2.2. Clinical evaluation
Pigs were evaluated 2 days prior to infection (−2 DPI) and daily
for 5 days after SIV infection for respiratory disease. Pigs were
observed for signs of respiratory disease including labored and/or
abdominal breathing and coughing both at rest and after obtain-
for 5 days after SIV infection.
Pigs were euthanized using a pentobarbital-based euthanasia
solution (Beuthanasia, Schering-Plough, Kenilworth, NJ, USA) fol-
lowed by exsanguination. The lungs were removed and evaluated
for pneumonia. Macroscopic lesions associated with SIV pneu-
monia, consisting of well demarcated dark-purplish areas of lung
consolidation , were sketched onto a standard lung diagram.
The proportion of lung surface with lesions was determined from
ously described . Bronchial swabs were obtained from each pig
Experimental design, vaccination status, SIV infection status and number of pigs.
GroupVaccinationSIV infectionTotal number of pigs
P. Kitikoon et al. / Vaccine 28 (2010) 523–531
ological procedures. Bronchoalveolar lavage (BAL) fluid collection
was performed by using 25ml of phosphate buffer saline (PBS)
with 100U/ml penicillin and 100mg/ml streptomycin as previ-
ously described . A portion of lung tissue was collected from
all lung lobes, fixed in 10% neutral buffered formalin processed
and embedded in paraffin using an automated tissue processor.
Lung sections were scored for microscopic lung lesions consistent
with SIV (necrotic bronchiolitis) as described previously . The
evaluation primarily focused on airway damage but also took into
The presence of SIV-specific antigen was assessed in the
formalin-fixed lung tissues using a previously described immuno-
histochemistry (IHC) staining method . IHC was performed on
sections cut from one paraffin-embedded lung tissue block and
included two pieces (1–2cm) of lung collected at necropsy.
2.4. Virus isolation
Following sample collection at −1, 2, 3, and 5 DPI, nasal swabs
and 1mg/ml trypsin). Viruses isolated from nasal swabs were
ulated onto MDCK cells followed by incubation at 37◦C with 5%
CO2as described previously . Virus was identified by staining
with anti-influenza A nucleoprotein monoclonal antibody (clone
HB-65, ATCC, Rockville, Maryland) followed by rabbit anti-mouse
teria, California) . The color was developed using a chromogen
aminoethyl carbazole substrate (Sigma, St. Louis, Missouri). Each
assay contained mock-infected negative control cells and positive
control cells infected at a known virus titer. The titer of the virus
in each nasal swab was expressed as log10 TCID50per milliliter
calculated by the method of Reed and Muench .
2.5. Hemagglutination-inhibition (HI) assay
The HI assays were performed according to a standard
protocol routinely used at ISU-Veterinary Diagnostic Labo-
ratory using 0.5% rooster erythrocytes for hemagglutination
. Virus antigens utilized in the HI assays included both
the challenge viruses, A/Swine/Iowa/15/1930 H1N1 (IA30) and
A/Swine/Minnesota/00194/2003 H1N2 (MN03).
2.6. ELISA for mucosal SIV-specific antibody production and
serum SIV M2e-specific antibody production
2.6.1. ELISA for mucosal SIV-specific antibody
BAL fluids were incubated at 37◦C for 1h with an equal amount
of 10mM dithiothreitol (DTT; Sigma–Aldrich, St. Louis, MO) to dis-
rupt mucus present in the fluids. ELISA assays for SIV antibodies in
the respiratory tract were performed as previously described .
Briefly, inactivated challenge virus was diluted to a hemaggluti-
nation (HA) concentration of 100 HA units/50?l. Immulon-2HB
96-well plates (Dynex, Chantilly, VA) were coated with 100?l of
SIV antigen and incubated at room temperature overnight. Plates
were blocked for 1h with 100ml of 10% BSA in PBS and washed 3
times with PBST washing buffer (0.1M phosphate buffer saline pH
7.2 with 0.02% Tween 20). The assay was performed on each BAL
sample in duplicate. Negative controls (DTT with equal amount of
room temperature for 1h and incubated with peroxidase-labeled
goat anti-swine IgG (Kirkegaard and Perry, Gaithersburg, MD) or
peroxidase-labeled goat anti-swine IgA (Bethyl, TX) at 37◦C for
1h. The ABTS/peroxidase was added as the substrate (Kirkegaard
and Perry, Gaithersburg, MD). Antibody levels were reported as
the mean optical density (OD) and the mean OD of each treatment
group was compared.
2.6.2. ELISA for serum SIV M2e-specific antibody
Serum samples collected at −29, −1 and 5 DPI were assayed
for SIV M2e-specific antibodies as described previously . In
brief, M2e coated plates were blocked with 100?l of 5% milk dilu-
ent (Kirkegaard & Perry Laboratories Inc., Gaithersburg, MD) for
1h at room temperature and washed 3 times with PBST wash-
ing buffer. Samples were diluted 1:50 using 5% milk diluent.
Each diluted sample was run in duplicate using 100?l of sample
per well and incubated for 1h at 37◦C. Excess antibodies were
removed by washing 3 times with PBST washing buffer. To detect
the presence of SIV M2e-specific antibodies, peroxidase-labeled
goat anti-swine immunoglobulin G (Sigma–Aldrich, St. Louis, MO)
was used and color was developed with a pre-warmed 2,2?-azino-
di-(3-ethylbenzthiazoline-6-sulfonate) (ABTS) substrate freshly
prepared according to the manufacturer’s protocol (Kirkegaard &
Perry Laboratories Inc., Gaithersburg, MD). The enzyme–substrate
reaction was stopped by adding 100?l of 1% sodium dodecyl sul-
was measured at 405nm. Positive samples were run in duplicates
on each plate and the sample diluent was included in the blank
wells as controls. The M2e-specific antibody levels were reported
as the mean OD of the duplicates.
2.7. Flow cytometry analysis
2.7.1. Culture procedures
Peripheral blood mononuclear cells (PBMC) were collected in
heparinized blood collection tubes (Becton, Dickinson and Com-
pany, Franklin Les, NJ) and isolated by differential centrifugation.
PBMCs were collected 1 day prior to challenge and at necropsy.
The PBMCs were counted prior to staining with carboxyfluorescein
2×107PBMCs were centrifuged (400×g) for 10min, supernatants
were aspirated, and cells were stained with 1× PBS pH 7.2 con-
taining 5?M CFSE. Cells were gently vortex for 5min. The staining
was quenched by adding 2ml of fetal bovine serum and incubated
for 2min to adsorb the dye. Cells were then washed three times
with RPMI 1640 (Mediatech, Huntingford, VA). Once stained, cells
(Costar, Corning, NY) at a density of 5×105cells per well in 100?l
medium (RPMI containing 10% fetal calf serum, 2mM l-glutamine,
tured for 4 days with inactivated IA30 and MN03 viruses (100 HA
units/100?l) and M2e antigen (10?g/100?l) in duplicate. Positive
control samples were cultured with 5mg/ml ConA in duplicate and
the culture media was used as negative control.
2.7.2. Cell surface marker staining
Stained PBMCs were centrifuged (400×g) for 10min and the
supernatant was discarded. Primary antibodies to swine leuko-
cyte surface antigens in PBS containing 1% BSA and 0.1% sodium
azide (FACS buffer) was added to wells containing cells. Primary
antibodies, including phycoerythrin (PE)-conjugated anti-CD4 and
biotinylated anti-CD8 were added to the appropriate wells. After
incubating for 20min, the cells were washed with FACS buffer
and resuspended in 50?l of secondary antibody streptavidin-
conjugated cychrome dye secondary antibody (Pharmingen, BD
Bioscience, CA). Cells were incubated, washed, resuspended and
fixed with 2% formalin in PBS before flow cytometric analysis.
The program Modfit Proliferation Wizard (Verity Software
House Inc., Topsham, Maine) was used to analyze cell prolifera-
P. Kitikoon et al. / Vaccine 28 (2010) 523–531
Fig. 1. Mean rectal temperature (≥40◦C) and clinical scores (0–3) for respiratory distress associated with SIV. Scoring: 0=normal; 1=mild dyspnea at rest; 2=moderate
dypsnea and/or tachypnea at rest; 3=severe dyspnea and tachypnea with distinct abdominal breathing.a,b,cValues not sharing the same letters are significantly different
tion. The results are presented as the mean number of proliferating
cells±standard error mean per 10,000 PBMCs. The number of cells
proliferating was calculated by the following formula: (% prolif-
eration to specific antigen×number of cells in the R1 gate)−(%
proliferation with no stimulation×number of cells in the R1 gate)
and side light scatter properties of porcine lymphocytes.
2.8. Statistical analysis
Analysis of variance (ANOVA) was performed to determine
group differences for each measured parameter. Significant dif-
ferences between treatment groups were evaluated using the
Tukey-Kramer Honestly Significant Difference multiple compari-
son test when P≤0.05. All data analyses were performed using
JMP®statistical software (SAS Institute, Cary, NC).
3.1. Clinical evaluation
Respiratory signs were mild in all pigs challenged with the
homologous IA30 virus (groups 2–5), independent of vaccination
status. Only the non-vaccinated, IA30-challenged pigs (group 5)
were febrile with rectal temperatures≥40◦C at 1 DPI (Fig. 1). Pigs
strated significantly increased respiratory disease levels compared
to the negative control pigs and the pigs challenged with IA30. Res-
piratory signs including cough and elevated respiratory rates were
significantly increased in the IA30 vaccinated pigs challenged with
MN03 with or without rM2 vaccination (groups 6 and 7) compared
to the rM2 vaccinated or non-vaccinated MN03-challenged pigs
(groups 8 and 9). Pigs vaccinated with the IA30 vaccine without
febrile for an average of 3 days, which was one day longer than pigs
vaccinated with only rM2 protein (group 7), although no statistical
difference was observed.
One pig that received only the rM2 vaccine prior to infec-
tion with IA30 virus died one day after SIV infection of unknown
and SIV-specific antigen was detected by immunohistochemistry
staining. No significant bacteria including M. hyopneumoniae were
cultured from the respiratory tract. The pig had no HI antibodies to
either viral antigens, but M2-specific IgG antibodies were detected
by the M2e-ELISA (data not shown).
3.2. Macroscopic and microscopic lesion scores and viral antigen
As shown in Table 2, the percentage of macroscopic lesions
consistent with SIV in the lungs of pigs that were vaccinated and
challenged with IA30 (groups 2–4) was significantly decreased
compared to the non-vaccinated group (group 5). The macro-
2–4) did not differ significantly from the non-vaccinated, non-
challenged pigs in group 1. Microscopic pneumonia scores were
IA30-challenged group was positive for SIV antigen.
Percentage of macroscopic lesions and microscopic lesion scores±standard error of the mean (S.E.M.) and swine influenza virus (SIV) antigen detection at 5 days post
infection (DPI). Means with different letters within columns are statistically different (P≤0.05).
Group VaccinationSIV infection% Macroscopic lesions†
Microscopic lesion scores‡
0.2 ± 0.1a
2.1 ± 1.1a
0.8 ± 0.3a
4.7 ± 0.9a,b
10.0 ± 2.5b
26.8 ± 3.3d
20.0 ± 2.1c,d
8.9 ± 0.7a,b
16.9 ± 2.3b,c
0.0 ± 0.0a
0.0 ± 0.0a
0.0 ± 0.0a
2.5 ± 0.6b,c
1.3 ± 0.6a,b
3.3 ± 0.4c
3.2 ± 0.2c
2.7 ± 0.6c
2.8 ± 0.4c
†As determined by lesion sketches and image analysis.
‡SIV microscopic lesion scores are based on the severity of bronchiolar epithelial damage (necrotic bronchiolitis).
P. Kitikoon et al. / Vaccine 28 (2010) 523–531
Fig. 2. Virus titers in nasal swabs collected from pigs at 2, 3 and 5 DPI. Results are represented as mean log10 TCID50/ml±S.E.M. Different letters within the same DPI are
significantly different (P≤0.05).
In contrast to the IA30 challenge groups, increased macroscopic
group 6 had significantly greater percentages of lung lesions com-
pared to the non-vaccinated pigs (group 9) and rM2 vaccinated
pigs (group 8). The enhanced disease observed in this study were
consistent with previous results observed with pigs immunized
with the IA30-vaccine and challenged with the MN03 virus .
Although not statistically significant when compared to groups 6
and 9, the addition of rM2 vaccine with the IA30 vaccine (group
7) appeared to have resulted in a slight reduction in the severity
of lung lesions, and pigs in group 8 that received only the rM2
vaccine had no evidence of vaccine enhancement of pneumonia.
There were no significant differences in the microscopic pneumo-
antigen was detected in the lungs of all the MN03-challenged pigs
at 5 DPI (Table 2). Interestingly, SIV antigen detected in the lungs
of IA30-vaccinated pigs (group 6) and non-vaccinated pigs (group
9) challenged with MN03 was not confined to the epithelial lining
of the large airways but were scattered in alveoli that were also
heavily infiltrated with mononuclear cells. In contrast, SIV antigen
detected in MN03-challenged pigs that received the rM2 vaccine
with IA30-vaccine (group 7) or rM2 vaccine alone (group 8) was
confined to the epithelial linings of large airways.
3.3. Virus levels in nasal secretions
No virus was isolated from the negative control pigs at any time
point in the study. Pigs challenged with IA30 virus shed low levels
of virus, with non-vaccinated pigs shedding the greatest amount,
In contrast, pigs in all groups challenged with MN03 virus (groups
6–9) shed equivalent levels of virus that were significantly greater
than the levels shed by the IA30-challenged pigs at 2 and 3 DPI
(Fig. 2). At 5 DPI no virus was detected in nasal secretions of MN03-
challenged pigs in the vaccinated groups 6 and 7. Virus was still
detected in nasal secretions from pigs that had received the rM2
vaccine or were not vaccinated (groups 8 and 9).
3.4. Hemagglutination-inhibition test
prior to vaccination, confirming that pigs were negative for SIV
viruses at the beginning of the study. Pigs in the negative control
Fig. 3. Mean hemagglutinin-inhibition (HI) antibody titers from IA30 challenged
pigs and -MN03-challenged pigs against the IA30 antigen (A) and (B) and the MN03
antigen (C) and (D). There were no positive HI titers against either virus antigen
in negative control pigs and pigs prior to vaccination. * Significantly higher values
observed at the same DPI (P≤0.05).
P. Kitikoon et al. / Vaccine 28 (2010) 523–531
Fig. 4. Mean O.D.±S.E.M. of SIV MN03-specific IgG (
virus (groups 2–5) measured by ELISA. Means with different letters within a column are statistically different (P≤0.05).
) and SIV MN03-specific IgA () from lower airways of negative control pigs (group 1) and pigs challenged with IA30
Fig. 5. Mean O.D.±S.E.M. of SIV MN03-specific IgG (
MN03 virus (groups 6–9) measured by ELISA. Means with different letters are statistically different (P≤0.05).
) and SIV MN03-specific IgA ( ) from lower airways of negative control pigs (group 1) and pigs challenged with
group remained seronegative throughout the study period (Fig. 3).
Prior to SIV challenge, HI antibody titers to the vaccine antigen
(IA30) were observed only in vaccinated pigs (groups 2, 3, 6 and
7). No cross-reactive HI antibody response to the MN03 antigen
was observed in sera from vaccinated pigs or IA30 challenged pigs.
In addition, no HI antibodies cross reacted with IA30 in the non-
vaccinated pigs challenged with the MN30 virus. However, the
level of HI antibodies to the MN03 antigen in IA30 vaccinated pigs
that had been challenged with the MN03 virus (groups 6 and 7)
were significantly higher compared to the non-vaccinated, MN03-
challenged pigs (group 9) at 5 DPI.
3.5.1. Local SIV-specific antibody response
While IgG antibodies that recognized the IA30 antigen in the
sera of the pigs challenged with MN03 were barely detectable by
the HI test at 5 DPI, antibodies that recognized both the MN03
and the IA30 viruses were detected in BAL fluid collected from
pigs that were only vaccinated with the IA30 antigen. Vaccinated
pigs that were infected with the IA30 virus had similar amounts of
cross-reacting IgG- and IgA-anti-MN03 antibodies in the BAL fluids
(Fig. 4). In contrast, pigs that were vaccinated and challenged with
MN03 virus had levels of cross-reacting IgG-anti MN03 antibod-
ies that were significantly higher than the cross-reacting IgA-anti
MN03 antibodies (Fig. 5).
3.5.2. Serum M2e-specific antibody response
Serum M2e-specific antibody responses were evaluated at 3
time points; prior to rM2 vaccine administration (−29 DPI), after
receiving 2 vaccinations (prior to SIV-challenge; −1 DPI) and 5 DPI.
No M2e-antibody levels were detected between the groups at −29
antibodies detected in pigs that received only the rM2 vaccine was
significantly higher compared to all other groups (Table 3). Serum
M2e-specific antibody responses were not significantly different
between the groups at 5 DPI.
3.6. Flow cytometry and cell surface marker analysis
Proliferation of different T cell populations in peripheral blood
to the recall antigens, IA30, MN03 and rM2 proteins, were studied
at two time points; one day (−1 DPI) prior to SIV infection, and
at 5 DPI. Since the treatment status prior to SIV infection between
IA30-vaccinated pigs from groups 2 and 6, IA30 and rM2 protein-
vaccinated pigs from groups 3 and 7 and rM2-vaccinated pigs from
groups 4 and 8 were considered the same, −1 DPI data retrieved
from these groups of pigs were combined for statistical analy-
sis. Significant differences in the number of T cells proliferating
in response to the recall antigens between the treatment groups
were detected only at the time point prior to SIV infection (Fig. 6).
Pigs that received the IA30-vaccine with or without rM2 vaccine
had significantly higher numbers of CD8+ T cells that proliferated
SIV-M2e IgG specific serum antibodies±standard error of the mean (S.E.M.). Means
with different letters within columns are statistically different (P≤0.05).
Groups Vaccination SIV infectionSerum IgG-M2e-specific antibody
0.077 ± 0.016a
0.122 ± 0.011a,b
0.262 ± 0.093a,b
0.449 ± 0.063b,c
0.086 ± 0.010a,b
0.156 ± 0.044a,b
0.265 ± 0.127a,b
0.652 ± 0.292c
0.088 ± 0.008a
0.123 ± 0.022a
0.145 ± 0.029a
0.189 ± 0.060a,b
0.207 ± 0.025a,b
0.140 ± 0.018a
0.133 ± 0.020a,b
0.325 ± 0.073b
0.125 ± 0.015a,b
0.162 ± 0.022a,b
P. Kitikoon et al. / Vaccine 28 (2010) 523–531
Fig. 6. Mean numbers of CD4+, CD8+ and CD4/8+ T cells±S.E.M. that proliferated to the IA30 (
cell surface marker staining. Peripheral blood mononuclear cells were collected from pigs two weeks after the second vaccination prior to SIV infection (−1 DPI). Means with
different letters compared between treatment groups within the same antigen are statistically different (P≤0.05).
), MN03 ( ) and rM2 ( ) antigen as determined by flow cytometry and
in response to both IA30 and MN03 antigen stimulation. Pigs that
received only the rM2 vaccine had significantly higher numbers
of CD8+ T cells that proliferated in response to the rM2 protein
compared to the negative control pigs or pigs vaccinated with only
IA30-vaccine. No significant difference in the numbers of the CD4+
and CD4/8+ T cells was detected between all of the groups. No sig-
nificant difference in the T cell populations was detected between
the groups at 5 DPI.
protein to protect against a H1N2 subtype SIV infection when used
in combination with an inactivated H1N1 SIV vaccine. This study
was performed using conventional pigs and a previously described
experimental H1N1-vaccine and heterologous H1N2 SIV-challenge
model . Pigs in this model have demonstrated enhanced disease
when vaccinated with an inactivated H1N1 (IA30) virus prior to
infection with the heterologous H1N2 (MN03) virus. In this study,
groups of pigs that received the rM2 protein were included to eval-
uate the efficacy of the rM2 protein alone to provide protection
against SIV infection.
Similar to the previous findings by Vincent et al. , increased
pneumonic lesions and SIV vaccine-associated disease were
heterologous SIV. The findings of this study further confirmed that
the immune response induced by an inactivated swine influenza
virus (H1N1) vaccine can have a negative impact when pigs are
infected with a heterologous swine influenza virus (H1N2). These
results underscore the possibility of similar scenario under field
conditions since the current commercial swine influenza vaccines
typically include older H1N1 and H3N2 viruses that may differ
genetically from the viruses that circulate in the swine population.
In this study, the addition of rM2 vaccine with the IA30-vaccine
P. Kitikoon et al. / Vaccine 28 (2010) 523–531
appeared to have helped reduce the number of febrile days by
one. In addition, the rM2 vaccine reduced macroscopic lung lesions
and the non-vaccinated, MN03-challenged group (16.9%). How-
ever, the rM2 vaccine administrated with the IA30-vaccine did not
reduce virus shedding significantly.
This study had two groups of pigs that were vaccinated with
only the rM2 vaccine and were then challenged with either the
IA30 or the MN03 viruses. Pigs vaccinated with rM2 and infected
with MN03 virus did not demonstrate the disease enhancement
observed in the IA30 vaccinated group. In fact, compared to the
IA30-vaccinated pigs, the rM2-vaccinated, MN03-challenged pigs
were febrile one day less, had significantly reduced respiratory
signs and the percentage of pneumonic lesions were not statisti-
cally different from the negative control pigs. One rM2-vaccinated
pig was found dead a day after infection with the H1N1 virus.
in epithelial linings of the large airways with no significant micro-
scopic lesions. While the cause of death of this pig was unknown,
an earlier study in pigs, demonstrated SIV vaccine-associated dis-
ease enhancement and death in pigs challenged with a H1N1 virus
following vaccination with nucleoprotein and M2 DNA-vaccine
. That study noted that the presence of a strong cell mediated
immune response towards NP and M2 protein alone without anti-
HA neutralizing antibodies may have been damaging to the host.
Pigs vaccinated with a purified M2 protein fused with hepatitis B
specific antibodies may have demonstrated SIV vaccine-associated
disease enhancement. Earlier findings indicate that M2-specific
antibodies are non-neutralizing but appear to contribute to virus
reduction through antibody-dependent natural killer cell activity
(ADCC) especially during the initial stage of infection when the
amount of virus is low . The expression of M2 protein on the
surface of SIV-infected cells in the presence of non-neutralizing
anti-M2 antibodies, but in the absence of anti-HA neutralizing
antibodies may induce cell membrane damage via ADCC and/or
complement fixation. The cause of death of the pig that was rM2-
So it is unknown if this enhanced immune response is a possible
cause of death of this single pig. However, no pigs in any other
group were impacted. Prior to SIV infection, pigs receiving only the
rM2 vaccine had significantly higher numbers of M2-specific CD8+
T cell in their PBMCs compared to IA30-vaccinated pigs with or
without rM2 vaccine. The rM2-vaccinated, MN03-challenged pigs
had reduced SIV-associated clinical signs and pneumonia, suggest-
ing a beneficial role of the M2 specific CD8+ T cells under these
conditions. The unexpected difference between these two groups
requires further examination in future studies to look at the mech-
anism of both the disease enhancement and the role an immune
response to the rM2 may play in reducing disease severity.
Of the antibodies produced in response to influenza infection,
HA antibodies are considered the most important as they are
primarily responsible for virus neutralization . Protective anti-
bodies against infection with SIV are based on both quality and
quantity of the HA antibodies produced. Studies in humans, mice
and ferrets suggests that serum anti-HA IgG antibodies transudate
into the bronchoalveolar passages and prevents the lower respira-
tory tract from being infected [34,35]. Clinical protection against
experimentally challenged SIV induced disease appears correlated
with the serum HI titer of the individual pig due to priming of the
immune system to the HA protein and the production of HI anti-
bodies that are antigenically similar to the infecting strain [36,37].
In addition, locally produced SIV-specific IgA antibodies found in
the mucosal lung areas are equally important for protecting the
local airways from influenza virus infection . These mucosal
SIV-specific antibodies in the nasal wash fluids and BAL fluids can
be detected as early as 4–5 days post primary SIV exposure [29,38]
and the predominant SIV-specific antibody isotype found in both
mucosal areas following primary infection is IgA, as opposed to
IgG, which is the highest isotype detected in serum . Studies
suggest that local IgA plays a significant role in defending against
SIV-induced disease as it provides heterosubtypic cross-protection
revealed up to 21% difference between the H1N1 (IA30) and H1N2
(MN03) viruses used in this vaccine-challenge model . In this
study, the serum IgG antibody response to the IA30 vaccine was
present at high levels in the vaccinated pigs prior to SIV chal-
by the HI assay prior to infection. Yet after infection, MN03-virus
specific IgG levels were significantly higher in the vaccinated pigs
compared to the non-vaccinated pigs. This indicates that the IA30
vaccine primed an antibody response that was cross-reactive to
the MN03 virus. The mucosal antibody response at 5 DPI from
IA30-vaccinated pigs contained high levels of IgG that were cross-
reactive with the MN03 antigen using a whole virus-based ELISA.
It is possible that the MN03-specific IgG antibodies detected in
the BAL fluids of the IA30-vaccinated, MN03-challenged pigs were
directed against a conserved area of the HA protein such as the
HA2 region and were of non-neutralizing nature. These findings
are supported by the presence of mucosal MN03 cross-reactive IgG
antibodies that did not reduce the level of virus shed in the early
phases of infection. In fact, the presence of the non-neutralizing
antibodies may have contributed to the potentiated lung lesions
found in the IA30-vaccinated, MN03-challenged pigs. The mucosal
MN03-IgA cross-reactive antibodies were also present with MN03-
IgG cross-reactive antibodies at 5 DPI indicating that the IA30
vaccine had primed the local IgA antibody response. Recent data
suggests specific anti-SIV-IgA antibodies have a broad spectrum
infections [3,39]. In this study, the IA30-vaccinated pigs challenged
with MN03 virus had significantly lower levels of anti-MN03 IgA
antibodies than IgG antibodies. It is possible that the level of
pre-existing IgA antibodies prior to MN03 infection were at an
insufficient level to overcome the disease enhancement caused by
the non-neutralizing IgG antibodies.
The administration of the rM2 vaccine in addition to the IA30
vaccine did not appear to reduce the HI antibody production to
the IA30 antigen. In contrast, the combined vaccination strategy
appeared to suppress the production of M2 specific antibodies as
the level was significantly lower in pigs receiving both vaccines
compared to pigs that received only the rM2 vaccine. Previous
data has shown that the influenza A proteins, HA and NA, when
presented as antigens on the same viral particle have an “intraviri-
onic antigenic competition” and the antibody production to the
HA protein out competes the NA protein . This phenomenon
disappears when the HA and NA proteins are administered sepa-
rately . This “antigenic competition” resulting in a reduced or
biased antibody response has been reported in combination vac-
cines studies [43–45]. Furthermore, it has been determined that
the M2 protein is poorly immunogenic in nature . Thus, it is
possible that the concentration of the rM2 protein administered
alone as a vaccine, or when used in combination with the IA30
vaccine was suboptimal for providing complete protection against
SIV infection. Future M2 vaccine studies are needed to determine if
optimizing the rM2 and HA protein concentration and ratio would
enhance protection against disease.
Other M2 protein-based vaccine trials have been conducted
in mice, monkeys and ferrets, which are not natural hosts for
influenza, found reduction in both clinical signs and virus excre-
tion. The findings in this study conducted in pigs, a natural host
P. Kitikoon et al. / Vaccine 28 (2010) 523–531
of influenza A viruses, indicates a potentially positive outcome in
using rM2 protein as a vaccine for SIV-associated disease preven-
tion. However, the results of this study underscore the importance
of inducing an immune response against the major envelope pro-
tein, HA to reduce SIV excretion. New innovative intervention
strategies, including vaccines consisting of various influenza A pro-
tein combinations, may be useful for the control of influenza A
viruses in the future.
The authors thank Dr. Cornelia Schroeder (Abteilung Virologie,
Institut für Mikrobiologie und Hygiene, Universität des Saarlandes,
Homburg, Germany) for her guidance in rM2 protein purification.
We also would like to thank Nancy Upchurch and the students in
the Thacker lab for their assistance in this project. This work was
supported by a grant from Iowa Livestock Health Advisory Council.
 Webster RG, Bean WJ, Gorman OT, Chambers TM, Kawaoka Y. Evolution and
ecology of influenza A viruses. Microbiol Rev 1992;56(1):152–79.
 Rimmelzwaan GF, Baars M, van Beek R, de Lijster P, de Jong JC, Claas EC, et al.
Influenza virus subtype cross-reactivities of haemagglutination inhibiting and
virus neutralising serum antibodies induced by infection or vaccination with
an ISCOM-based vaccine. Vaccine 1999;17(20–21):2512–6.
 Tamura S, Tanimoto T, Kurata T. Mechanisms of broad cross-protection pro-
vided by influenza virus infection and their application to vaccines. Jpn J Infect
for vaccination policy. Vaccine 1999;17(11–12):1321–8.
 Webby RJ, Rossow K, Erickson G, Sims Y, Webster R. Multiple lineages of anti-
genically and genetically diverse influenza A virus co-circulate in the United
States swine population. Virus Res 2004;103(1–2):67–73.
 Van Reeth K, Van Gucht S, Pensaert M. Investigations of the efficacy of Euro-
pean H1N1- and H3N2-based swine influenza vaccines against the novel H1N2
subtype. Vet Rec 2003;153(1):9–13.
 Vincent AL, Lager KM, Janke BH, Gramer MR, Richt JA. Failure of protection and
enhanced pneumonia with a US H1N2 swine influenza virus in pigs vaccinated
with an inactivated classical swine H1N1 vaccine. Vet Microbiol 2007.
 Rimmelzwaan GF, Osterhaus AD. Universal influenza vaccine still far in the
future. Ned Tijdschr Geneeskd 1999;143(46):2305–7.
 De Filette M, Min Jou W, Birkett A, Lyons K, Schultz B, Tonkyro A, et al. Uni-
versal influenza A vaccine: optimization of M2-based constructs. Virology
 Ito T, Gorman OT, Kawaoka Y, Bean WJ, Webster RG. Evolutionary analysis of
the influenza A virus M gene with comparison of the M1 and M2 proteins. J
 Pinto LH, Holsinger LJ, Lamb RA. Influenza virus M2 protein has ion channel
activity. Cell 1992;69(3):517–28.
 Zebedee SL, Richardson CD, Lamb RA. Characterization of the influenza virus
cloned cDNA. J Virol 1985;56(2):502–11.
 Liu W, Li H, Chen YH. N-terminus of M2 protein could induce antibodies with
inhibitory activity against influenza virus replication. FEMS Immunol Med
 Frace AM, Klimov AI, Rowe T, Black RA, Katz JM. Modified M2 proteins produce
 Neirynck S, Deroo T, Saelens X, Vanlandschoot P, Jou WM, Fiers W. A universal
influenza A vaccine based on the extracellular domain of the M2 protein. Nat
 Slepushkin VA, Katz JM, Black RA, Gamble WC, Rota PA, Cox NJ. Protection
of mice against influenza A virus challenge by vaccination with baculovirus-
expressed M2 protein. Vaccine 1995;13(15):1399–402.
 Mozdzanowska K, Feng J, Eid M, Kragol G, Cudic M, Otvos Jr L, et al. Induction of
influenza type A virus-specific resistance by immunization of mice with a syn-
thetic multiple antigenic peptide vaccine that contains ectodomains of matrix
protein 2. Vaccine 2003;21(19–20):2616–26.
 Fan J, Liang X, Horton MS, Perry HC, Citron MP, Heidecker GJ, et al. Preclinical
study of influenza virus A M2 peptide conjugate vaccines in mice, ferrets, and
rhesus monkeys. Vaccine 2004;22(23–24):2993–3003.
 Heinen PP, de Boer-Luijtze EA, Bianchi AT. Respiratory and systemic humoral
and cellular immune responses of pigs to a heterosubtypic influenza A virus
infection. J Gen Virol 2001;82(Pt 11):2697–707.
 Heinen PP, Rijsewijk FA, de Boer-Luijtze EA, Bianchi AT. Vaccination of pigs
with a DNA construct expressing an influenza virus M2-nucleoprotein fusion
protein exacerbates disease after challenge with influenza A virus. J Gen Virol
 Kitikoon P, Strait EL, Thacker EL. The antibody responses to swine influenza
virus (SIV) recombinant matrix 1 (rM1), matrix 2 (M2), and hemagglu-
tinin (HA) proteins in pigs with different SIV exposure. Vet Microbiol
 Janke BH. Classical swine influenza. Large Anim Pract 1998;19:24–9.
 Thacker EL, Thacker BJ, Janke BH. Interaction between Mycoplasma hyopneu-
moniae and swine influenza virus. J Clin Microbiol 2001;39(7):2525–30.
 Mengeling WL, Lager KM, Vorwald AC. Diagnosis of porcine reproductive and
respiratory syndrome. J Vet Diagn Invest 1995;7(1):3–16.
 Vincent LL, Janke BH, Paul PS, Halbur PG. A monoclonal-antibody-based
immunohistochemical method for the detection of swine influenza virus in
 Kitikoon P, Nilubol D, Erickson BJ, Janke BH, Hoover TC, Sornsen SA, et al.
The immune response and maternal antibody interference to a heterologous
H1N1 swine influenza virus infection following vaccination. Vet Immunol
 Reed LJ, Muench H. A simple method of estimating fifty percent endpoints. Am
J Hyg 1938;27:493–7.
 Yoon KJ, Janke BH, Swalla RW, Erickson G. Comparison of a commercial H1N1
enzyme-linked immunosorbent assay and hemagglutination inhibition test in
detecting serum antibody against swine influenza viruses. J Vet Diagn Invest
 Larsen DL, Karasin A, Zuckermann F, Olsen CW. Systemic and mucosal
immune responses to H1N1 influenza virus infection in pigs. Vet Microbiol
 Dumitriu IE, Mohr W, Kolowos W, Kern P, Kalden JR, Herrmann M. 5,6-
as well as detergent-treated cells can be traced in composite cell samples. Anal
of swine lymphocytes for detection of proliferation, apoptosis, viability, and
phenotype. Cytometry 2002;48(3):146–52.
 Jegerlehner A, Schmitz N, Storni T, Bachmann MF. Influenza A vaccine based
on the extracellular domain of M2: weak protection mediated via antibody-
dependent NK cell activity. J Immunol 2004;172(9):5598–605.
Dev Biol (Basel) 2003;115:25–30.
 Greenbaum E, Furst A, Kiderman A, Stewart B, Levy R, Schlesinger M, et al.
a single intra-nasal immunization with a new inactivated trivalent influenza
vaccine. Vaccine 2002;20(7–8):1232–9.
 Renegar KB. Influenza virus infections and immunity: a review of human and
animal models. Lab Anim Sci 1992;42(3):222–32.
 de Jong JC, Palache AM, Beyer WE, Rimmelzwaan GF, Boon AC, Osterhaus
AD. Haemagglutination-inhibiting antibody to influenza virus. Dev Biol (Basel)
 Heinen PP, van Nieuwstadt AP, Pol JM, de Boer-Luijtze EA, van Oirschot JT,
Bianchi AT. Systemic and mucosal isotype-specific antibody responses in pigs
to experimental influenza virus infection. Viral Immunol 2000;13(2):237–47.
 Lee BW, Bey RF, Baarsch MJ, Larson ME. Class specific antibody response to
influenza A H1N1 infection in swine. Vet Microbiol 1995;43(2–3):241–50.
 Tamura S, Funato H, Hirabayashi Y, Suzuki Y, Nagamine T, Aizawa C, et al.
Cross-protection against influenza A virus infection by passively transferred
respiratory tract IgA antibodies to different hemagglutinin molecules. Eur J
 Vincent AL, Lager KM, Ma W, Lekcharoensuk P, Gramer MR, Loiacono C, et al.
States. Vet Microbiol 2006;118(3–4):212–22.
 Johansson BE, Bucher DJ, Kilbourne ED. Purified influenza virus hemagglutinin
and neuraminidase are equivalent in stimulation of antibody response but
induce contrasting types of immunity to infection. J Virol 1989;63(3):1239–46.
 Johansson BE, Kilbourne ED. Dissociation of influenza virus hemagglutinin
and neuraminidase eliminates their intravirionic antigenic competition. J Virol
 Braun R, Babiuk LA, van Drunen Littel-van den H. Compatibility of plasmids
expressing different antigens in a single DNA vaccine formulation. J Gen Virol
 Frey S, Dagan R, Ashur Y, Chen XQ, Ibarra J, Kollaritsch H, et al. Interference of
antibody production to hepatitis B surface antigen in a combination hepatitis
A/hepatitis B vaccine. J Infect Dis 1999;180(6):2018–22.
 Romano M, Roupie V, Hamard M, Huygen K. Evaluation of the immuno-
genicity of pBudCE4.1 plasmids encoding mycolyl-transferase Ag85A and
phosphate transport receptor PstS-3 from Mycobacterium tuberculosis. Vac-
 Feng J, Zhang M, Mozdzanowska K, Zharikova D, Hoff H, Wunner W, et al.
Influenza A virus infection engenders a poor antibody response against the
ectodomain of matrix protein 2. Virol J 2006;3:102.