Vaccine 34 (2016) 55125518

Contents lists available at ScienceDirect

Vaccine
journal homepage: www.elsevier.com/locate/vaccine

Mutations in the haemagglutinin protein and their effect in transmission

of highly pathogenic avian influenza (HPAI) H5N1 virus in sub-optimally
vaccinated chickens
Ioannis Sitaras a,b,, Xanthoula Rousou a, Ben Peeters b, Mart C.M. de Jong a,
a
b

Quantitative Veterinary Epidemiology, Department of Animal Sciences, Wageningen University, Radix Building 107, Droevendaalsesteeg 1, 6708 PB Wageningen, The Netherlands
Department of Virology, Central Veterinary Institute of Wageningen University and Research Centre, Houtribweg 39, 8221 RA Lelystad, The Netherlands

a r t i c l e

i n f o

a b s t r a c t

Article history:
Received 19 July 2016
Received in revised form 29 September
2016
Accepted 3 October 2016
Available online 8 October 2016
Keywords:
Mutations
HA
Vaccination
Transmission
H5N1
Influenza
Epidemiology

Background: Transmission of highly pathogenic avian influenza (HPAI) viruses in poultry flocks is associated with huge economic losses, culling of millions of birds, as well as human infections and deaths. In
the cases where vaccination against avian influenza is used as a control measure, it has been found to
be ineffective in preventing transmission of field strains. Reports suggest that one of the reasons for this
is the use of vaccine doses much lower than the ones recommended by the manufacturer, resulting in
very low levels of immunity. In a previous study, we selected for immune escape mutants using homologous polyclonal sera and used them as vaccines in transmission experiments. We concluded that provided a threshold of immunity is reached, antigenic distance between vaccine and challenge strains
due to selection need not result in vaccine escape. Here, we evaluate the effect that the mutations in
the haemagglutinin protein of our most antigenically-distant mutant may have in the transmission efficiency of this mutant to chickens vaccinated against the parent strain, under sub-optimal vaccination
conditions resembling those often found in the field.
Methods: In this study we employed reverse genetics techniques and transmission experiments to examine if the HA mutations of our most antigenically-distant mutant affect its efficiency to transmit to vaccinated chickens. In addition, we simulated sub-optimal vaccination conditions in the field, by using a
very low vaccine dose.
Results: We find that the mutations in the HA protein of our most antigenically-distant mutant are not
enough to allow it to evade even low levels of vaccination-induced immunity.
Discussion: Our results suggest that for the antigenic distances we investigated vaccination can
reduce transmission of an antigenically-distant strain compared to the unvaccinated groups, even when
low vaccine doses are used, resulting in low levels of immunity.
2016 Elsevier Ltd. All rights reserved.

1. Introduction
A particular lineage of highly pathogenic avian influenza (HPAI)
H5N1 is unique among other HPAI strains in that since its emergence in Guangdong province in China in 1996, it has spread on
a global scale [1]. Its efficiency in transmitting within and between
poultry flocks, coupled to severe morbidity and high (up to 100%)
case mortality rates, make it a serious threat to the poultry industry. Culling of infected poultry or pre-emptive culling is one of the
Corresponding authors at: Quantitative Veterinary Epidemiology, Department
of Animal Sciences, Wageningen University, Radix Building 107, Droevendaalsesteeg 1, 6708 PB Wageningen, The Netherlands (I. Sitaras).
E-mail addresses: ioannis.sitaras@wur.nl (I. Sitaras), mart.dejong@wur.nl (M.C.
M. de Jong).

main means to control infection and results in huge economic

losses, as well as ethical dilemmas. In addition, with this HPAI
H5N1 lineage, human infections are often reported. To this date,
the World Health Organisation (WHO) reports 851 cases of human
infections with HPAI H5N1 (mainly among poultry workers and
their families), out of which 450 were fatal [2,3]. Although
human-to-human transmission is still not reported, many believe
that HPAI H5N1 virus is one of the prime candidates for a new pandemic influenza outbreak [4].
One of the most important genes in the influenza genome is the
haemagglutinin (HA) gene, encoding for the HA protein. HA is the
major surface protein of influenza viruses, outnumbering the neuraminidase (NA) protein by a ratio of 4:1 [5,6]. It is essential for
influenza virulence, transmissibility and antigenicity, since it is

http://dx.doi.org/10.1016/j.vaccine.2016.10.002
0264-410X/ 2016 Elsevier Ltd. All rights reserved.

Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 14, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

I. Sitaras et al. / Vaccine 34 (2016) 55125518

the primary target of neutralising antibodies [513]. Indeed, the

HA experiences the most intense selection pressure compared to
other parts of the genome and is particularly prone to mutations
at both the nucleotide and amino acid level due to the errorprone RNA polymerase of the virus [6,9,11,1416]. As a consequence, mutants that are antigenically different from their parent
strains are often selected in the field, and this is thought to be
one of the reasons why vaccines are failing to protect against transmission [5,6,17].
Although vaccination against HPAI H5N1 is used in some countries, transmission still takes place and in some instances vaccination has been implicated in the selection of antigenic variants
capable of escaping vaccination-induced immunity [4,1720].
Oftentimes, the failure of vaccination to stop transmission has
been attributed to the antigenic distances between vaccine and
field strains, thus requiring constant vaccine updating (similar to
vaccination against human influenza). This has led many to believe
that vaccination of poultry is not cost-effective and should not be
used as a means to stop transmission of field strains. However,
recent reports suggest that vaccine inefficiency in stopping transmission could be attributed to its failure in inducing sufficiently
high levels of immune response even against the vaccine strain
itself, to a sufficiently high percentage of the vaccinated population
[4,17,18,2124].
In previous studies we have selected for escape mutants of the
HPAI H5N1 A/turkey/Turkey/1/2005 virus (from now on abbreviated to H5N1 t/T) using homologous polyclonal sera and have characterised these mutants genetically and antigenically [25]. We
then studied the effect of vaccination-induced immunity and antigenic distance on the transmission dynamics of the parent strains
to animals vaccinated with the mutants, by using either a high
(128 Haemagglutination Units, HAU) or low (4 HAU) vaccine dose
[26]. We concluded that vaccine dose (and consequent immunity)
has a much bigger effect on transmission than the antigenic distances between the vaccine and challenge strains we studied. In
addition, we calculated the minimum level of immunity (P8
Haemagglutination Inhibition Units, HIU, as measured by HI titres),
as well as the minimum percentage of the vaccinated population
(P86.5%) necessary to display said immunity, in order for transmission to be prevented (R < 1).
In this study, we examine the effect that the mutations found in
the HA protein of our latest and most antigenically-distant mutant
(t/T-P42) may have on the transmission of this mutant to suboptimally vaccinated animals. For this reason, we used reverse
genetic techniques to insert the HA of t/T-P42 into the backbone
of its H5N1 t/T parent strain, thus creating strain rgt/T-P42. We
then used rgt/T-P42 to challenge animals vaccinated with the
H5N1 t/T parent strain, so as to simulate a field situation in which
such an antigenically-distant strain may well be the co-circulating
field strain, selected under immune pressure. To examine if there
are differences in the transmission efficiency between the parent
strain and the rgt/T-P42 mutant, we also included a homologous
group, in which animals were similarly sub-optimally vaccinated
and then challenged with the parent strain. Sub-optimal field vaccination was simulated by using a very low vaccine dose of 1 HAU,
resulting in very low levels of immunity of approximately but not
higher than 8 HIU (68 HIU) and high percentages of animals
exhibiting said low immunity (P86.5% of vaccinated population).
This would theoretically lead to R < 1 in the homologous group
but a high probability for some contact infections (as a minor outbreak in this group). We analysed the results from the transmission
experiments, using Generalised Linear Models (GLM) in order to
see whether the HA mutations result in a higher transmission rate
parameter of the mutant compared to the same parameter estimated for the parent strain in a vaccinated population.

5513

2. Materials and methods

2.1. Construction of reverse genetics viruses
The HA genes from the parent strain H5N1 t/T and the t/T-P42
mutant were synthesised by GenScript (Piscataway, NJ) and
inserted into the pHW2000 bi-directional transcription plasmid
using BsmBI cloning sites [27]. The sequence of the HA gene constructs was verified by means of sequencing the HA gene as previously described [25]. The sequence of the HA gene of the t/T-P42
mutant is submitted to GenBank (GenBank: KF042153).
The purified pHW2000-derived plasmids containing the HA of
the H5N1 t/T or t/T-P42 mutants and the remaining 7 genes of
the H5N1 t/T parent strain were used to rescue viruses using
reverse genetics. The detailed protocol is described in the supporting information file.
2.2. Pilot vaccination experiment to determine optimum vaccination
dose
All animals used in this experiment were specific pathogen-free
(SPF) white leghorn chickens, derived from SPF embryonated
chicken eggs (ECEs) (Charles River Avian Vaccine Services) that
were hatched and raised in our HCU facilities. Inactivation of
viruses for vaccine construction took place as previously described
[25,26].
In previous work [26] we have calculated and described the
threshold level of immunity as expressed in HI titres necessary
to stop transmission of HPAI H5N1 strains in vaccinated chickens
(under experimental conditions). This threshold was calculated
to be P8 HI units (HIU) to be reached by at least 86.5% of animals
at three weeks post-vaccination (p.v.). In this study, we aim to use
a sub-optimal vaccination dose that would lead to levels of immunity of 68 HIU, as measured with the homologous (vaccine) virus,
thus maximising the chance that HI titres would be much lower in
the heterologous case.
In order to ensure that vaccinated animals would have HI titres
of 68 HIU at three weeks p.v., we performed a pilot experiment, in
which we vaccinated three groups, each comprising 7 SPF 3-weekold chickens, with three different vaccine doses (1, 2 and 4 HAU) of
the inactivated reverse genetics parent strain (rgH5N1 t/T) in the
presence of adjuvant (Stimune, Prionics), at a 4:5 volume/volume
(v/v) inactivated virus-to-Stimune ratio. The vaccine was delivered
intra-muscularly (i.m.) in the leg muscle with 0.5 mL of the respective vaccine dose. The animals were housed in the same room but
in separate cages for each group, and were checked twice a day for
mortality and signs of disease. At 2 and 3 weeks p.v., 2 mL of blood
were taken from each chicken, sera were prepared and inactivated
at 56 C for 50 mins. The sera were used in HI assays, in which their
HI titre against the vaccine strain was quantified. HI assays were
performed in duplicate as previously described [25,26,28,29].
2.3. Transmission experiment
For the transmission experiment, seventy new SPF chickens
were vaccinated i.m. in the leg muscle at 3 weeks of age with
0.5 mL of 1 HAU of inactivated rgH5N1 t/T in the presence of adjuvant (Stimune, Prionics) at a 4:5 (v/v) inactivated virus-to-Stimune
ratio. The same procedure was followed for another 20 chickens
belonging to the two unvaccinated control groups (groups 3 and
4), only these animals were injected i.m. with 0.5 mL of allantoic
fluid from SPF ECEs in the presence of adjuvant as described above.
One animal belonging to one of the unvaccinated groups (Group 3)
died as a result of a shock while the mock vaccine was being
administered and was removed from the group. All animals were

Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 14, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

5514

I. Sitaras et al. / Vaccine 34 (2016) 55125518

checked twice a day for mortality and signs of disease. Two weeks
post-vaccination (p.v.), 2 mL of blood were taken from each of the
70 vaccinated animals, sera were prepared and inactivated and
their HI titres were quantified in HI assays using the vaccine strain
as antigen. From the results of the HI assays and by extrapolating
the expected HI titres at three weeks p.v. based on the pilot vaccination experiment, 20 chickens with expected titres of 68 HIU
against the vaccine strain at 3 weeks p.v. were selected from the
initial 70, and randomly assigned to the two vaccinated groups
(Groups 1 and 2). The reason we initially vaccinated 70 chickens
and we selected 20 out of the 70 for the transmission experiment
was that the vaccine dose used (1 HAU) was so low that inevitably
a large number of chickens would have undetectable HI titres at
3 weeks p.v., so we wanted to maximise the number of chickens
with detectable HI titres. In addition, the reason we had to choose
these 20 chickens by extrapolating their HI titres against the vaccine strain at 3 weeks p.v. was that challenge of these chickens
was to take place immediately after collection of blood at 3 weeks
p.v., therefore the time frame did not allow for the results from the
HI assays to be available before challenge.
In total, four groups of animals were used. Each group consisted
of ten 3-week-old SPF chickens, of which 5 would be inoculated
(designated I) and 5 would be contacts (designated S for susceptible). Two of the groups were the ones vaccinated with rgH5N1 t/T
and challenged with either the homologous strain or rgt/T-P42
(Groups 1 and 2 respectively). Two additional groups served as
controls, consisting of unvaccinated chickens challenged with
either of the two strains (Groups 3 and 4). Therefore, the four
groups consisted of the following:
Group 1: rgH5N1 t/T vaccinated & rgH5N1 t/T challenged
(homologous group for comparing transmission efficiency
between parent strain and mutant in a sub-optimally vaccinated population).
Group 2: rgH5N1 t/T vaccinated & rgt/T-P42 challenged
(heterologous group for comparing transmission efficiency
between parent strain and mutant in a sub-optimally vaccinated population).
Group 3: Unvaccinated & rgH5N1 t/T challenged (serving as a
control for transmission of rgH5N1 t/T in an unvaccinated
population).
Group 4: Unvaccinated & rgt/T-P42 challenged (serving as a
control for transmission of rgt/T-P42 in an unvaccinated
population).
At three weeks p.v. 2 mL of blood were taken from each of the
20 vaccinated chickens and the chickens were housed separately
according to their assigned group. On the same day, 5 animals/group were inoculated with 105 TCID50/0.2 mL (0.1 mL intranasally and 0.1 mL intra-tracheally) of the rgH5N1 t/T (Groups 1
and 3) or rgt/T-P42 (Groups 2 and 4) viruses and the 5 contact animals/group were added 24 h post-inoculation. TCID50 assays were
performed using standard protocols. Briefly, a monolayer of
Madin-Darby canine kidney (MDCK) cells was infected with 10fold virus dilutions in 96-well plates. After a 48 h incubation, an
immunoperoxidase monolayer assay (IPMA) was performed. The
results of the staining were evaluated under the microscope and
the TCID50 titre was calculated using the Reed-Muench formula.
From there on the experiment was carried out as previously
described [26], with tracheal and cloacal swabs taken at days 1
7, 10 and 14 post-inoculation (p.i.), animals checked twice a day
for mortality and signs of disease (including during the days when
routine sampling was not taken) and additional cloacal swabs
taken from deceased animals before they were removed from their
group. Each swab was placed in tubes containing 2 mL of tryptose

phosphate broth at the time of sampling. The swabs were stored in

-80 C pending further analysis.
At the end of the experiment (14 days p.i.) 5 mL of blood were
collected from all surviving animals. The animals were then euthanized, concluding the experiment. Viral RNA was isolated from all
swabs and quantified by means of qPCR as previously described
[26]. Briefly, the qPCR and AI probe used were optimised to be
specific to the highly conserved M gene of influenza A viruses. In
total, 45 cycles of amplification were used.
2.4. Statistical analysis and mathematical model
The Reproduction ratio (R) [30] was used as a measure for the
transmission between individuals. R was estimated based on the
SIR stochastic model. The infection rate parameters (b) for the different groups, as well as their variances and 95% confidence intervals (C.I.), were estimated using generalised linear models (GLM)
and the statistical software STATA (Version 10). Furthermore, the
average infectious periods (AIP) of all infected animals were also
calculated, as well as their variances and 95% C.I. [31,32]. For the
estimation or R, the natural logarithms (ln) of the transmission rate
parameter and the AIP were added, to obtain ln(R) and its confidence interval. All 95% C.I. are reported at the normal scale, but
as they were estimated at the ln scale, they are not symmetrical.
We consider differences statistically significant if the p-value of
these differences compared to 0 is below the threshold of 0.05.
Our mathematical model is described in detail in [26] and a
detailed description is also found in the supporting information.
3. Results
3.1. Determination of optimum vaccination dose
The results from the pilot experiment showed that in order to
obtain animals with HI titres as close to 8 HIU as possible at
3 weeks p.v., the lowest vaccination dose (1 HAU) would have to
be used. Supplementary Table S1 shows the HI assay results from
all three vaccine doses at 2 and 3 weeks p.v.
3.2. Transmission experiment
HI assays performed using sera from all vaccinated animals
(inoculated and contacts) and cross-checked against the vaccine
and challenge strains rgH5N1 t/T and rgt/T-P42 showed either
undetectable or very low titres at 3 weeks p.v. The exception was
one contact animal in Group 1 which showed a HI titre of 128
HIU against the vaccine strain (Table 1). Times of death and animals found to be qPCR-positive for influenza virus in tracheal
and cloacal swabs are shown in Table 1.
In the unvaccinated control groups 3 and 4 (unvaccinated/
rgH5N1 t/T challenged and unvaccinated/rgt/T-P42 challenged,
respectively), all contact animals became infected by day 3 p.i.
and died, except one contact animal in Group 3, which survived
the infection and recovered by day 7 p.i (Table 1). As a consequence, both groups had a similarly high R0, which was calculated
to be 3.65 and 3.05 for groups 3 and 4 respectively (Table 2). Both
groups exhibited comparably short AIP (1.58 and 1.81 day
respectively).
In Group 1 (rgH5N1 t/T vaccinated/rgH5N1 t/T challenged), 4
animals (one inoculated and three contacts) had detectable HI
titres against the challenge strain at 3 weeks p.v. In this group,
all inoculated animals were dead by day 5 p.i. Transmission was
observed in 4/5 contact animals, three of which died by day 8 p.
i., whilst one recovered from infection on day 8 p.i. (Table 1). By
contrast, no transmission was observed in Group 2 (rgH5N1 t/T

Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 14, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

5515

I. Sitaras et al. / Vaccine 34 (2016) 55125518

Table 1
Table showing an overview of transmission experiment results from Day 1 p.i until Day 14 p.i. HI titres are based on sera collected at 3 weeks p.v. (immediately before challenge).
Transmission was observed in both unvaccinated control groups and in the homologous rgH5N1 t/T group but not in the rgH5N1 t/T vaccinated/rgt/T-P42 challenged group. Grey
boxes indicate days when animals were considered to be positive for influenza. S: Susceptible; I: Infectious; NA: Not Applicable; +/ or
/ when positive or negative tracheal
sample; /+ or /
when positive or negative cloacal sample; y: Dead Animal. Cells with diagonal lines represent days when samples were not routinely taken, except cloacal
samples from deceased or euthanized animals. Animals were checked twice a day throughout the duration of the experiment.

Table 2
Transmission dynamics of each group of animals and of combined vaccinated or unvaccinated groups. NA: Not Applicable.
Group no.

Vaccine strain

Challenge strain

b (day
1) (95% CI)

Var (b)

Infectious period (day) (95% CI)

R (95% CI)

VarR

1
2
3
4
1&2
3&4

rgH5N1 t/T
rgH5N1 t/T
Unvaccinated
Unvaccinated
Vaccinated
Unvaccinated

rgH5N1 t/T
rgt/T-P42
rgH5N1 t/T
rgt/T-P42

0.52
0
2.43
1.68
0.19
7.77

1.29
NA
1.26
1.23
1.29
1.58

2.21
3.68
1.58
1.81
2.65
1.70

1.14 (0.225.93)
0
3.82 (0.8517.28)
3.05 (0.8411.05)
0.50 (0.092.88)
13.19 (2.4770.52)

2.02
NA
1.81
1.54
2.23
2.08

(0.191.38)
(0.956.22)
(0.694.13)
(0.070.50)
(2.0729.17)

vaccinated/rgt/T-P42 challenged), whilst all inoculated animals

were found to be qPCR-positive for influenza and 4/5 were dead
by day 10 p.i. (Table 1). In this group, two inoculated and one contact animal showed detectable HI titres to the challenge strain at
3 weeks p.v. The progress of infection of the contact animals in
all four groups is shown in Fig. 1. The Ct values of all positive animals are shown in Supplementary Tables S2-S5.
3.3. Quantification of transmission (SIR model)
For each of the 4 groups, the transmission rate parameter (b)
was estimated as ln(b) based on the observed infectious and
susceptible animals at the start of each time interval and the number of new cases during that time interval.

(0.598.29)
(0.7318.57)
(0.485.12)
(0.724.56)
(0.6211.40)
(0.614.76)

The estimates from the GLM analysis for each group are shown
in Table 2 and the dataset used for this analysis is shown in Supplementary Table S6.
In addition to calculating b, AIP and R for each group, they were
also calculated for both vaccinated (combined) and both unvaccinated (combined) groups (Table 2). This was done in order to
demonstrate that the vaccinated groups had a lower R than the
unvaccinated ones, and that R < 1 after vaccination, when taking
into account heterogeneity in circulating viruses. Thus, the overall
estimate of b and of the AIP in vaccinated populations were 0.19 (C.
I.: 0.070.50) day
1 and 2.65 (C.I.: 0.6211.40) day respectively.
The overall calculated estimate of R in vaccinated populations
was 0.50 (C.I.: 0.092.88). In unvaccinated populations the overall
estimate of b and of the AIP were 7.77 (C.I.: 2.0729.17) day
1 and

Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 14, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

5516

I. Sitaras et al. / Vaccine 34 (2016) 55125518

Fig. 1. Observed progress of contact animal infections. Progress of infection of contact animals (according to qPCR-positive trachea and cloaca swabs) per group per day.
Contact animals were added on Day 1 p.i., immediately after sampling (trachea and cloaca swabs) of inoculated animals took place. Group 1: rgH5N1 t/T vaccinated/rgH5N1 t/
T challenged; Group 2: rgH5N1 t/T vaccinated/rgt/T-P42 challenged; Group 3: Unvaccinated/rgH5N1 t/T challenged; Group 4: Unvaccinated/rgt/T-P42 challenged.

1.70 (C.I.: 0.614.76) day respectively. The overall calculated estimate of R0 in unvaccinated populations was 13.19 (C.I.: 2.47
70.52). For the comparison between combined vaccinated and
combined unvaccinated groups, we added the dummy variable
VAC (vaccinated), which has a value of 1 when vaccinated and a
value of 0 when not vaccinated. We estimated the vaccination
effect to be
2.39 (C.I.:
3.56 to
1.22), with a p-value of 0.000.
The results show (not significantly) that the R for Group 1 is
above 1 but lower than the R0 for the unvaccinated Group 3. The
R for Group 2 was 0 since no transmission occurred in this group.
The infectivity rate of animals belonging to the two vaccinated
groups (Groups 1 and 2) is significantly lower than the one of animals belonging to the unvaccinated groups (Groups 3 and 4). In
addition, animals in the vaccinated groups had higher AIPs than
animals in the unvaccinated groups (Table 2).
4. Discussion
In this study we examined whether the mutations found in the
HA protein of our most antigenically-distant mutant (t/T-P42)
allow it to transmit better than the parent strain to animals vaccinated with the parent strain. We use this mutant as a challenge
strain since it may well be a potentially co-circulating field strain,
selected under vaccination-induced immune pressure. Additionally, we attempt to simulate sub-optimal vaccine coverage in the
field by using a very low vaccine dose (1 HAU), resulting in high
percentages of animals exhibiting low levels of vaccinationinduced immune response.
In the unvaccinated control groups 3 and 4 all contact animals
became infected and mortality was 90% and 100% respectively.
This is in agreement with the highly pathogenic status of the challenge viruses and their high transmission rates in unvaccinated
individuals [33,34]. The Ct values measured from tracheal and
cloacal swabs indicated a high virus load (Supplementary Tables
S4 and S5). In both groups, the average infectious period was very
short and the estimated R0 was high and taken together it was
significantly above the threshold value of 1 (Table 2). Given the
high mortality rate and the short infectious period, both challenge
viruses used could rapidly cause a major outbreak in a population
consisting of immunologically-nave chickens, at least under
experimental conditions.
In the homologous Group 1 (rgH5N1 t/T vaccinated/rgH5N1 t/T
challenged), 4/5 contact animals became infected and the AIP was

short. Upon examination of the HI titres of the contact animals in

this group, infection and death correlated to low or undetectable
HI titres to the challenge strain. The same observations apply to
the inoculated animals of this group. As a consequence of the
low HI titres and the highly pathogenic nature of the challenge
strain, all inoculated animals died early on in the experiment
(Table 1). The Ct values obtained from tracheal and cloacal swabs
indicated a high virus load. In this group, the R was calculated to
be above the threshold value of 1 (1.14) (Table 2). Although this
group should in theory have exhibited the highest level of protection against transmission (since the vaccine and challenge strain
were the same), instead transmission was observed. This could
be explained by the observation that since only one of five inoculated animals had a detectable HI titre against the vaccine strain
(which was the same as the challenge strain), these animals were
clearly infectious, which is also plausibly demonstrated by their
low Ct values.
In Group 2 (rgH5N1 t/T vaccinated/rgt/T-P42 challenged), no
transmission was observed. In total there were two inoculated
and one contact animal with detectable HI titres against the challenge strain. Due to the level of vaccination-induced immunity
being very low in this group, it would be expected that rgt/T-P42
would transmit. Since it didnt, it could be that even these low
levels of immunity could still prevent transmission of rgt/T-P42.
Thus, it would appear that in this group, although animals are
not protected against inoculation they are protected against transmission. Similar results have been reported elsewhere. In a study
evaluating the effect of vaccination with H7N1 on the transmission
of H7N7, transmission of the latter was reduced even though the
level of the immune response was very low [35]. The AIP in Group
2 (3.68 days) is longer than that of Group 1. The chances of having
a minor outbreak in Group 2 are higher than in Group 1, since the
inoculated animals in Group 2 although they survived longer
are less infectious. Since there were no infections of contact animals, the R in this group was calculated to be 0 (Table 2). In addition, the fact that in this group two inoculated animals had
detectable HI titres to the challenge strain compared to only one
such animal in Group 1 and that the Ct values of the inoculated
animals were much higher making them less infectious, further
explains why transmission was not seen in Group 2.
The inability of the rgt/T-P42 to transmit to a vaccinated population cannot be attributed to the mutations it carries in the HA
having a detrimental effect in transmissibility or pathogenicity,

Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 14, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

I. Sitaras et al. / Vaccine 34 (2016) 55125518

since rgt/T-P42 behaves similarly to rgH5N1 t/T in an unvaccinated

population. Besides, we have previously reported that the HA
mutations in the t/T-P42 strain are located in sites of antigenic
importance, have also been found in strains selected in the field
and are thought to confer a fitness advantage [25]. Therefore, the
fact that rgt/T-P42 did not transmit to a vaccinated population
even under favourable conditions, indicates that the antigenic distance between itself and the challenge strain as reported in [25],
although the biggest distance that could probably be obtained with
the type of selection used, is not large enough for rgt/T-P42 to
escape even low levels of vaccination-induced immunity. Indeed,
in our previous study [26] in which transmission of the parent
strain H5N1 t/T was evaluated in chickens vaccinated with a higher
dose (4 HIU) of a vaccine made from the t/T-P42 mutant, the challenge strain H5N1 t/T was also not able to transmit. In that case,
the higher vaccine dose resulted in all animals exhibiting immune
titres against the challenge strain averaging 50 HIU, further
demonstrating that the antigenic distance between these two
strains (4.5 HIU or 22.17 as calculated in [25]), although comparable
to distances found in the field [36,37], is not enough for the challenge strain to evade vaccination-induced immunity [26]. In that
same study [26], we demonstrated that the antigenic distance
between vaccine and challenge strains plays a much smaller role
than vaccine dose (and consequent immunity) in the transmission
of HPAI H5N1 strains to vaccinated animals, since a higher vaccine
dose results in a higher percentage of animals with HI titres P8
HIU. These results are corroborated in the present study, since
transmission was seen in the homologous Group 1, where the antigenic distance between vaccine and challenge strain is 0. A similar
observation was seen in [26], in which transmission was also seen
in one of the homologous groups, due to the lower percentage of
animals with HI titres P8 HIU. There is a chance that the HA mutations reported in the t/T-P42 mutant may have resulted in lowering the virulence of this strain without increasing its
transmission ability. This could give the t/T-P42 an advantage in
the longer term, since the longer AIP demonstrated in Group 2
could mean that animals stay infected in the herd for longer before
showing clinical signs of disease. In addition, possible mutations in
other parts of the t/T-P42 genome may be able to increase the
transmission efficiency of this mutant, and this could be analysed
in future studies.
Finally, it is possible that the lack of transmission of rgt/T-P42
could be a random event, since the dataset was relatively small
and no replicates were performed, thus not allowing us to conclude
that R is indeed below 1 and calculate the probability for a minor
outbreak more accurately. However, given the outcome, it is not
possible that R > 1 significantly for this group, even if the experiment would have been replicated. In contrast, in the homologous
scenario (Group 1), rgH5N1 t/T was able to transmit in a vaccinated
population, resulting in R > 1, indicating that although a very low
vaccination-induced immune response may not protect against
transmission, it can reduce the R compared to unvaccinated
groups.
We support that it is not necessarily the escape of the virus that
causes transmission in the field, but also the fact that oftentimes
the vaccines have failed to induce immunity in animals even
against the originally-circulating field virus. Further evidence for
this also comes from field vaccination studies that show that the
levels of herd immunity achieved were very low and the vaccinated animals exhibited very low or undetectable levels of immunity, even to the vaccine seed strain [3840]. As a consequence,
vaccination failed to prevent transmission not only of
antigenically-different strains, but also of the matching circulating
field strains. Other studies have shown that the effect of antigenic
distance in transmission could be overcome by inducing sufficient
levels of protection in vaccinated animals [17,2224,3335,4143].

5517

In addition, the mutations that may be selected in these escape

mutants, do not necessarily make them transmit better than the
original circulating strains (with which vaccines are usually
matched).
The results of our study indicate that vaccination can reduce
transmission of an antigenically-distant strain compared to the
unvaccinated groups, even when low vaccine doses are used,
resulting in low levels of immunity. As a consequence, we believe
that future efforts should focus on increasing the effectiveness of
vaccination and should use HI titres against the challenge strains
at the start of a vaccination campaign as indicators of a vaccines
potential to stop transmission.
5. Ethics statement
All animal experiments complied fully with Dutch and European Union Law (2010/63/EU) and were reviewed and approved
by the Dierexperimenten Commissie (DEC) Animal Sciences Group,
Lelystad (animal experiments ethical committee) prior to being
carried out (permit numbers 2,012,070 and 2,012,071). All animals
were housed and handled in accordance to European Union Directive 2010/63/EU on the protection of animals used for scientific
purposes, in a way that promoted natural behaviours including
social interaction, foraging and exercise. Water and food were provided ad libitum. All animals were housed separately according to
group (i.e. one group/room). Provisions were made in the protocols
to ameliorate animal suffering, such as regular monitoring of animal facilities, healthcare of animals and euthanasia of animals at
the end of the experiment (or during the experiment if serious
health problems were noticed).
All animal experiments were performed within the High Containment Unit (HCU), in Bio-Safety Level 3+(BSL3+) facilities at
the Central Veterinary Institute of Wageningen UR.
Conflicts of interest
None.
Acknowledgements
The authors would like to acknowledge Olav de Leeuw for his
insight and help with the construction of the reverse genetics
mutants and Prof. Ron Fouchier for kindly providing the H5N1
backbone. This research was funded by the Economic Structure
Enhancement Fund (FES) in The Netherlands: FES Program on
Avian Influenza. The funding source had no involvement in the
study design, collection, analysis and interpretation of data, writing of the report, or in the decision to submit the article for
publication.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.vaccine.2016.10.
002.
References
[1] Xu X, Subbarao, Cox NJ, Guo Y. Genetic characterization of the pathogenic
influenza A/Goose/Guangdong/1/96 (H5N1) virus: similarity of its
hemagglutinin gene to those of H5N1 viruses from the 1997 outbreaks in
Hong Kong. Virology 1999;261:159.
[2] Peiris JS, de Jong MD, Guan Y. Avian influenza virus (H5N1): a threat to human
health. Clin Microbiol Rev 2007;20:24367.
[3] WHO. Cumulative number of confirmed human cases for avian influenza A
(H5N1) reported to WHO, 20032016, <http://www.who.int/influenza/
human_animal_interface/2016_06_13_tableH5N1.pdf?ua=1>; 2016.

Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 14, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.

5518

I. Sitaras et al. / Vaccine 34 (2016) 55125518

[4] Smith GJ, Fan XH, Wang J, Li KS, Qin K, Zhang JX, et al. Emergence and
predominance of an H5N1 influenza variant in China. Proc Natl Acad Sci USA
2006;103:1693641.
[5] Bouvier NM, Palese P. The biology of influenza viruses. Vaccine 2008;26(Suppl
4):D4953.
[6] Webster RG, Bean WJ, Gorman OT, Chambers TM, Kawaoka Y. Evolution and
ecology of influenza A viruses. Microbiol Rev 1992;56:15279.
[7] Philpott M, Hioe C, Sheerar M, Hinshaw VS. Hemagglutinin mutations related
to attenuation and altered cell tropism of a virulent avian influenza A virus. J
Virol 1990;64:29417.
[8] Webster RG, Rott R. Influenza virus A pathogenicity: the pivotal role of
hemagglutinin. Cell 1987;50:6656.
[9] Kao CL, Chan TC, Tsai CH, Chu KY, Chuang SF, Lee CC, et al. Emerged HA and NA
mutants of the pandemic influenza H1N1 viruses with increasing
epidemiological significance in Taipei and Kaohsiung, Taiwan, 200910. PLoS
ONE 2012;7:e31162.
[10] Stevens J, Blixt O, Tumpey TM, Taubenberger JK, Paulson JC, Wilson IA.
Structure and receptor specificity of the hemagglutinin from an H5N1
influenza virus. Science 2006;312:40410.
[11] Wilson IA, Cox NJ. Structural basis of immune recognition of influenza virus
hemagglutinin. Annu Rev Immunol 1990;8:73771.
[12] Steinhauer DA. Role of hemagglutinin cleavage for the pathogenicity of
influenza virus. Virology 1999;258:120.
[13] Rott R. The pathogenic determinant of influenza virus. Vet Microbiol
1992;33:30310.
[14] Steinhauer DA, Holland JJ. Rapid evolution of RNA viruses. Annu Rev Microbiol
1987;41:40933.
[15] Holland J, Spindler K, Horodyski F, Grabau E, Nichol S, VandePol S. Rapid
evolution of RNA genomes. Science 1982;215:157785.
[16] Nelson MI, Holmes EC. The evolution of epidemic influenza. Nat Rev Genet
2007;8:196205.
[17] Cattoli G, Milani A, Temperton N, Zecchin B, Buratin A, Molesti E, et al.
Antigenic drift in H5N1 avian influenza virus in poultry is driven by mutations
in major antigenic sites of the hemagglutinin molecule analogous to those for
human influenza virus. J Virol 2011;85:871824.
[18] Cattoli G, Fusaro A, Monne I, Coven F, Joannis T, El-Hamid HS, et al. Evidence
for differing evolutionary dynamics of A/H5N1 viruses among countries
applying or not applying avian influenza vaccination in poultry. Vaccine
2011;29:936875.
[19] Balish AL, Davis CT, Saad MD, El-Sayed N, Esmat H, Tjaden JA, et al. Antigenic
and genetic diversity of highly pathogenic avian influenza A (H5N1) viruses
isolated in Egypt. Avian Dis 2010;54:32934.
[20] Lee CW, Senne DA, Suarez DL. Effect of vaccine use in the evolution of Mexican
lineage H5N2 avian influenza virus. J Virol 2004;78:837281.
[21] UN FAO. H5N1 HPAI global overview JanuaryMarch 2012; 2012.
[22] Peyre M, Samaha H, Makonnen YJ, Saad A, Abd-Elnabi A, Galal S, et al. Avian
influenza vaccination in Egypt: limitations of the current strategy. J Mol Genet
Med: Int J Biomed Res 2009;3:198204.
[23] Domenech J, Dauphin G, Rushton J, McGrane J, Lubroth J, Tripodi A, et al.
Experiences with vaccination in countries endemically infected with highly
pathogenic avian influenza: the Food and Agriculture Organization
perspective. Rev Sci Tech 2009;28:293305.
[24] Soares Magalhaes RJ, Pfeiffer DU, Otte J. Evaluating the control of HPAIV H5N1
in Vietnam: virus transmission within infected flocks reported before and after
vaccination. BMC Veter Res 2010;6:31.
[25] Sitaras I, Kalthoff D, Beer M, Peeters B, de Jong MC. Immune escape mutants of
highly pathogenic avian influenza H5N1 selected using polyclonal sera:
identification of key amino acids in the HA protein. PLoS ONE 2014;9:e84628.

[26] Sitaras I, Rousou X, Kalthoff D, Beer M, Peeters B, de Jong MC. Role of

vaccination-induced immunity and antigenic distance in the transmission
dynamics of highly pathogenic avian influenza H5N1. J Roy Soc, Interf/Roy Soc
2016;13.
[27] 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 USA 2000;97:610813.
[28] Donald HB, Isaacs A. Counts of influenza virus particles. J Gen Microbiol
1954;10:45764.
[29] UN OIE. Manual of diagnostic tests and vaccines for terrestrial animals
(Mammals, Birds and Bees). 7th ed.; 2012.
[30] Diekmann O, Heesterbeek JA, Metz JA. On the definition and the computation
of the basic reproduction ratio R0 in models for infectious diseases in
heterogeneous populations. J Math Biol 1990;28:36582.
[31] Velthuis AG, Bouma A, Katsma WE, Nodelijk G, De Jong MC. Design and
analysis of small-scale transmission experiments with animals. Epidemiol
Infect 2007;135:20217.
[32] Klinkenberg D, de Bree J, Laevens H, de Jong MC. Within- and between-pen
transmission of Classical Swine Fever Virus: a new method to estimate the
basic reproduction ratio from transmission experiments. Epidemiol Infect
2002;128:2939.
[33] Bouma A, Claassen I, Natih K, Klinkenberg D, Donnelly CA, Koch G, et al.
Estimation of transmission parameters of H5N1 avian influenza virus in
chickens. PLoS Pathog 2009;5:e1000281.
[34] Poetri ON, Bouma A, Murtini S, Claassen I, Koch G, Soejoedono RD, et al. An
inactivated H5N2 vaccine reduces transmission of highly pathogenic H5N1
avian influenza virus among native chickens. Vaccine 2009;27:28649.
[35] van der Goot JA, Koch G, de Jong MC, van Boven M. Quantification of the effect
of vaccination on transmission of avian influenza (H7N7) in chickens. Proc Natl
Acad Sci USA 2005;102:181416.
[36] Koel BF, van der Vliet S, Burke DF, Bestebroer TM, Bharoto EE, Yasa IW, et al.
Antigenic variation of clade 2.1 H5N1 virus is determined by a few amino acid
substitutions immediately adjacent to the receptor binding site. MBio 2014;5:
e01070114.
[37] Abbas MA, Spackman E, Fouchier R, Smith D, Ahmed Z, Siddique N, et al. H7
avian influenza virus vaccines protect chickens against challenge with
antigenically diverse isolates. Vaccine 2011;29:74249.
[38] Capua I, Cattoli G. Prevention and control of highly pathogenic avian influenza
with particular reference to H5N1. Virus Res 2013;178:11420.
[39] Hafez MH, Arafa A, Abdelwhab EM, Selim A, Khoulosy SG, Hassan MK, et al.
Avian influenza H5N1 virus infections in vaccinated commercial and backyard
poultry in Egypt. Poult Sci 2010;89:160913.
[40] Desvaux S, Grosbois V, Pham TT, Dao DT, Nguyen TD, Fenwick S, et al.
Evaluation of the vaccination efficacy against H5N1 in domestic poultry in the
Red River Delta in Vietnam. Epidemiol Infect 2013;141:77688.
[41] van der Goot JA, van Boven M, Stegeman A, van de Water SG, de Jong MC, Koch
G. Transmission of highly pathogenic avian influenza H5N1 virus in Pekin
ducks is significantly reduced by a genetically distant H5N2 vaccine. Virology
2008;382:917.
[42] Rauw F, Palya V, Van Borm S, Welby S, Tatar-Kis T, Gardin Y, et al. Further
evidence of antigenic drift and protective efficacy afforded by a recombinant
HVT-H5 vaccine against challenge with two antigenically divergent Egyptian
clade 2.2.1 HPAI H5N1 strains. Vaccine 2011;29:2590600.
[43] Terregino C, Toffan A, Cilloni F, Monne I, Bertoli E, Castellanos L, et al.
Evaluation of the protection induced by avian influenza vaccines containing a
1994 Mexican H5N2 LPAI seed strain against a 2008 Egyptian H5N1 HPAI virus
belonging to clade 2.2.1 by means of serological and in vivo tests. Avian
Pathol: J WVPA 2010;39:21522.

Downloaded from ClinicalKey.com at ClinicalKey Global Guest Users November 14, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.