﻿Evidence of Infection by H5N2 Highly Pathogenic Avian
Influenza Viruses in Healthy Wild Waterfowl
Nicolas Gaidet1.
*, Giovanni Cattoli2.
, Saliha Hammoumi1
, Scott H. Newman3
, Ward Hagemeijer4
, John Y.
Takekawa5
, Julien Cappelle1
, Tim Dodman4
, Tony Joannis6
, Patricia Gil1
, Isabella Monne2
, Alice Fusaro2
,
Ilaria Capua2
, Shiiwuua Manu7
, Pierfrancesco Micheloni8
, Ulf Ottosson9
, John H. Mshelbwala10
, Juan
Lubroth3
, Joseph Domenech3
, Franc
¸ois Monicat1
1 Centre de Coope
´ration Internationale en Recherche Agronomique pour le De
´veloppement, Montpellier, France, 2 Istituto Zooprofilattico Sperimentale delle Venezie,
Legnaro, Italy, 3 Food and Agriculture Organization of the United Nations, Animal Production and Health Division, Rome, Italy, 4 Wetlands International, Wageningen, The
Netherlands, 5 U.S. Geological Survey, Western Ecological Research Center, Vallejo California, United States of America, 6 National Veterinary Research Institute, Vom,
Nigeria, 7 AP Leventis Ornithological Research Institute, Jos, Nigeria, 8 Istituto Nazionale per la Fauna Selvatica, Bologna, Italy, 9 Ottenby Bird Observatory, Kehlen,
Luxembourg, 10 Federal Department of Forestry, Abuja, Nigeria
Abstract
The potential existence of a wild bird reservoir for highly pathogenic avian influenza (HPAI) has been recently questioned by
the spread and the persisting circulation of H5N1 HPAI viruses, responsible for concurrent outbreaks in migratory and
domestic birds over Asia, Europe, and Africa. During a large-scale surveillance programme over Eastern Europe, the Middle
East, and Africa, we detected avian influenza viruses of H5N2 subtype with a highly pathogenic (HP) viral genotype in
healthy birds of two wild waterfowl species sampled in Nigeria. We monitored the survival and regional movements of one
of the infected birds through satellite telemetry, providing a rare evidence of a non-lethal natural infection by an HP viral
genotype in wild birds. Phylogenetic analysis of the H5N2 viruses revealed close genetic relationships with H5 viruses of low
pathogenicity circulating in Eurasian wild and domestic ducks. In addition, genetic analysis did not reveal known
gallinaceous poultry adaptive mutations, suggesting that the emergence of HP strains could have taken place in either wild
or domestic ducks or in non-gallinaceous species. The presence of coexisting but genetically distinguishable avian influenza
viruses with an HP viral genotype in two cohabiting species of wild waterfowl, with evidence of non-lethal infection at least
in one species and without evidence of prior extensive circulation of the virus in domestic poultry, suggest that some strains
with a potential high pathogenicity for poultry could be maintained in a community of wild waterfowl.
Citation: Gaidet N, Cattoli G, Hammoumi S, Newman SH, Hagemeijer W, et al. (2008) Evidence of Infection by H5N2 Highly Pathogenic Avian Influenza Viruses in
Healthy Wild Waterfowl. PLoS Pathog 4(8): e1000127. doi:10.1371/journal.ppat.1000127
Editor: Ron A. M. Fouchier, Erasmus Medical Center, Netherlands
Received March 10, 2008; Accepted July 16, 2008; Published August 15, 2008
This is an open-access article distributed under the terms of the Creative Commons Public Domain declaration which stipulates that, once placed in the public
domain, this work may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose.
Funding: This extensive survey has been implemented in the framework of an FAO Technical Cooperation Programme, and has been made possible by
additional financial resources received from the governments of France and Sweden. We also acknowledge the Global Avian Influenza Network for Surveillance
(GAINS) program, funded in part by USAID Grant No. LAG-A-00-99-00047-00, for the larger joint wild bird surveillance activities in Nigeria. The opinions expressed
herein are those of the authors and do not necessarily reflect the views of the U.S. Agency for International Development.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: nicolas.gaidet@cirad.fr
. These authors contributed equally to this work.
Introduction
Wild waterbirds are considered the natural reservoirs of low
pathogenic avian influenza (LPAI) viruses [1,2], but highly
pathogenic avian influenza (HPAI) viruses responsible for high
mortality in domestic birds do not have recognised wild bird
reservoirs [3]. Among the 16 hemagglutinin (HA) subtypes of
avian influenza viruses (AIVs) perpetuated in wild birds, where
they caused unapparent or mild disease [1,4], H5 and H7 viruses
are recognised to have the potential to become highly pathogenic
(HP) in poultry. HPAI viruses are generally considered to emerge
from LPAI precursors once introduced and adapted to gallina-
ceous poultry populations, and to not occur in wild birds [5,6].
Prior to 2002, no HPAI virus had been isolated from free-living
waterbird populations, with the exception of a large mortality
event in Common Terns Sterna hirundo in South Africa in 1961
associated with an H5N3 HPAI infection [7], and a few isolated
cases in terrestrial birds associated with AIV-infected poultry flocks
[6]. The ecology of HPAI viruses has changed since 2002 with the
re-emergence and spread of the Asian H5N1 HPAI virus that has
been responsible for mortalities in more than 75 wild bird species
in 38 countries [8]. The spread of the H5N1 HPAI virus over Asia,
Europe and Africa, contemporary to the isolation of the virus in
dead migratory birds, questioned the potential for wild birds to
perpetuate and spread HPAI viruses.
In response to the inter-continental spread of the H5N1 HPAI
virus, we sampled and tested for HPAI infection in live-caught and
hunted wild birds (.11,000 birds of 144 species) in 19 countries over
Eastern Europe, the Middle East and Africa during 2006 [9] and
2007. In addition, we equipped some of the wild ducks (n= 45) we
caught with satellite transmitters to track their local and migratory
movements in relation to the potential spread of avian diseases.
During this large-scale surveillance programme, we detected the
presence of AIVs with an HP viral genotype (i.e. motif at the HA
PLoS Pathogens | www.plospathogens.org 1 August 2008 | Volume 4 | Issue 8 | e1000127
cleavage site consistent with highly pathogenic avian influenza) in
two wild African waterfowl species in Nigeria. One of the infected
ducks had been fitted with a satellite transmitter during a
concurrent satellite telemetry survey conducted in Nigeria,
allowing its movements to be monitored. Here we present
combined results from the molecular analysis of the viral genome
and the satellite tracking survey of wild waterfowl naturally
infected by HPAI viruses.
Results
Wild bird surveillance for HPAI
AIVs of the H5N2 subtype were detected by means of
molecular tests in free living and apparently healthy White-faced
Whistling Duck (WFWD) Dendrocygna viduata and Spur-winged
Goose (SWG) Plectropterus gambensis. These birds had been sampled
at the same lake in the Hadejia-Nguru Wetlands in northern
Nigeria (Jigawa State; latitude: 12u489N; longitude: 10u449E;
Figure 1), respectively on the 14th
(WFWD) and 17th
(SWG)
February 2007. In details, the oro-pharyngeal swab of one WFWD
out of nine (11.1%) tested positive for type A influenza viruses and
H5 subtype by RRT-PCR (Table 1). AIVs were also detected in
fresh faecal samples from 8 SWG out of 97 (8.2%), of which six
(6.2%) tested H5 positive. H5 positive samples showed clear
positive fluorescence signal, with real time PCR Cycle threshold
(Ct) values ranging from 26 to 30. Other waterbirds species
Author Summary
Until recently, the highly pathogenic avian influenza (HPAI)
viruses responsible for high mortality in some domestic
poultry were considered not to have a wild bird reservoir,
but to emerge in domestic poultry populations from low
pathogenic viruses perpetuated in wild waterbirds. The
rapid spread of H5N1 HPAI virus in 2005­2006, with
concurrent outbreaks reported in both domestic and wild
birds over Asia, Europe, and Africa, has raised concerns
about the potential role of migratory birds in the
epidemiology of the HPAI infection. Wild birds were
sampled in Africa and tested by molecular and virological
methods in an attempt to trace the circulation of HPAI
viruses. In addition, some of these wild birds were
equipped with satellite transmitters to track their local
and migratory movements in relation to the potential
spread of avian diseases. Avian influenza viruses (H5N2)
were detected in wild waterfowl in Nigeria, and were
subsequently characterized as highly pathogenic by
molecular sequencing (HPAI viral genotype). Movements
of one infected bird tracked by satellite telemetry revealed
that it survived infection by an HP viral genotype. This
result constitutes a rare finding of infection by an AIV with
an HPAI viral genotype in healthy wild birds.
Figure 1. Movement paths of one White-faced Whistling Duck (Dendrocygna viduata) fitted with a satellite transmitter, from Hadejia-
Nguru Wetlands in northern Nigeria to Lake Chad in western Chad, during February­March 2007. Dates and minimum distances
between main staging areas (orange circles) are indicated on the satellite image.
doi:10.1371/journal.ppat.1000127.g001
Highly Pathogenic Avian Influenza in Wild Birds
PLoS Pathogens | www.plospathogens.org 2 August 2008 | Volume 4 | Issue 8 | e1000127
(n = 122 birds), mostly waders, sampled at the same site during the
same period (1­17th
February 2007) all tested negative (Table 1).
Sequence analysis was able to reveal the presence of HPAI viral
genotypes, with multiple basic amino acid motif at the cleavage site
of the HA molecule (PQKEKRRKKR*GLF and PQREKRR-
KKR*GLF, Table 2) in four samples collected in four distinct birds
(one WFWD and three SWG, Table 1). Conventional RT-PCR
assay targeting the neuraminidase (NA) gene segment tested positive
for the N2 subtype. No virus isolate could be obtained but sequence
analysis of the entire segments encoding HA and NA proteins from
PCR products confirmed the presence of AIVs of H5N2 subtype.
Insufficient material of the five remaining samples was available for
sequencing and further analysis.
Molecular and phylogenetic analysis
The site where birds were sampled is situated in Northern
Nigeria where the first outbreak of H5N1 HPAI virus in Africa
was reported in January 2006 [10], with many subsequent
outbreaks in gallinaceous poultry or free-ranging ducks (the closest
past outbreak occurred 80 km away). Unexpectedly, phylogenetic
analysis of the entire HA and NA gene segments (accession
numbers EU544242 to EU544248) showed that the viruses were
unrelated to any strains of H5N1 HPAI viruses. Phylogenetic
analysis based on the HA gene clustered the sequences with
contemporary LPAI viral strains isolated in South and Central
Europe and with H5 viruses isolated in South Africa (Figure 2).
Nigerian H5 sequences revealed the highest homologies with the
H5N2 LPAI isolate A/mallard/Bavaria/1/2005 (98.2% for A/
SWG/Nigeria/5388-2-8-5/2007 and 97.9% for A/WFWD/
Nigeria/3927-1/2007). Analysis of the HA deduced amino acid
sequence of the Nigerian viruses A/WFWD/Nigeria/3927-1/
2007 and A/SWG/Nigeria/5388-2-8-5/2007 showed high simi-
larity with isolate A/mallard/Bavaria/1/2005, with only 11, 8 and
9 amino acid differences located outside the cleavage site,
respectively. High homology at the nucleotide level, ranging from
97% to 98%, was also revealed when the sequences were
compared to H5 LPAI viruses circulating in Southern Europe,
A/teal/Italy/3931-38/2005 (H5N2), A/teal/Italy/3931/2005
(H5N2), A/teal/Italy/3812/2005 (H5N3) and A/mallard/Italy/
5366/2007 (H5N2). Representative isolates of HPAI and LPAI
H5N2 viruses recently detected in South Africa, A/ostrich/
SouthAfrica/AI1091/2006 (HPAI) and A/ostrich/SouthAfrica/
AI1160/2006 (LPAI), also showed high nucleotide sequence
similarities (97.6%) to A/mallard/Bavaria/1/2005, but lower
homologies with the Nigerian HPAI H5N2 sequences (range
95.9%­96.4%). As shown in Figure 2, A/SWG/Nigeria/5388-2-
8-5/2007 and A/WFWD/Nigeria/3927-1/2007 sequences were
not closely related genetically to the other H5 HPAI viruses
previously detected in gallinaceous poultry in Southern Europe
(e.g. A/chicken/Italy/8/98 H5N2) and in Common Tern in
South Africa (A/tern/South Africa/1/61 H5N3). Phylogenetic
analysis based on the NA gene revealed similarities with LPAI
viruses isolated in Far East Asia (Figure 3). The highest homology
(98.3%­98.5%) was observed with A/duck/Jiang Xi/1286/2005
(H5N2). The N2 sequences of A/ostrich/South Africa/AI1091/
2006 (HPAI) and A/ostrich/South Africa/AI1160/2006 (LPAI)
[11] were not closely related, showing homologies ranging from
89%­89.5% (Figure 3).
Genetic markers supposed to be related to gallinaceous poultry
adaptation and virulence, such as potential additional glycosyla-
Table 1. Wild bird species sampled and tested for AIVs in Nigeria in 2007, and results of molecular tests and sequence analyses.
Bird group Species No. birds No. Type A positivea
/No. Samples
No. H5/No.
Type A HPAIb
Cloaca Oropharynx Faeces
Afro-tropical ducks White-faced Whistling-Duck Dendrocygna viduata 9 0/2 1/2 0/7 1/1 1
Spur-winged Goose Plectropterus gambensis 97 8/97 6/8 3
Comb Duck Sarkidiornis melanotos 8 0/7 0/8
Eurasian duckc
Garganey Anas querquedula 9 0/9 0/8
Eurasian waders Ruff Philomachus pugnax 49 0/11 0/11 0/38
Little Stint Calidris minuta 9 0/9 0/8
Wood Sandpiper Tringa glareola 1 0/1 0/1
Marsh Sandpiper Tringa stagnatilis 1 0/1 0/1
Jack Snipe Lymnocryptes minimus 1 0/1 0/1
Black-winged Stiltd
Himantopus himantopus 1 0/1 0/1
Afro-tropical waders Spur-winged Lapwing Vanellus spinosus 18 0/18 0/18
Greater Painted-Snipe Rostratula benghalensis 2 0/2 0/2
Herons Common Squacco Heron Ardeola ralloides 4 0/4 0/4
Yellow-billed Egret Egretta intermedia 1 0/1 0/1
Black-crowned Night-Heron Nycticorax nycticorax 1 0/1 0/1
Passerines Yellow Wagtail Motacilla flava 16 0/16 0/16
Raptors Pallid Harrier Circus macrourus 1 0/1 0/1
Total 17 species 228 0/85 1/84 8/142 7 4
a
Evaluated by means of real-time RT-PCR specific for type A influenza viruses (M gene).
b
Evaluated by genomic sequencing.
c
Birds that breed in the Western Palearctic and migrate to Africa.
d
Potentially either an Afro-tropical breeding or Eurasian wintering individual.
doi:10.1371/journal.ppat.1000127.t001
Highly Pathogenic Avian Influenza in Wild Birds
PLoS Pathogens | www.plospathogens.org 3 August 2008 | Volume 4 | Issue 8 | e1000127
tion sites (AGS) in the HA molecule [12] and stalk deletion in the
NA molecule [13], were not detected in the Nigerian sequences. In
the NA molecule, one potential additional glycosylation site was
detected in position 331 in A/WFWD/Nigeria/3927-1/2007
(S331N mutation) and in position 329 in A/SWG/Nigeria/5388-
8/2007 and A/SWG/Nigeria/5388-2/2007 (D329N mutation).
Sequence analysis also revealed that two distinct H5N2 viruses
with an HP viral genotype were present in these two cohabiting
species of the waterfowl community. Eight and seven amino acid
differences located outside the cleavage site were detected in A/
WFWD/Nigeria/3927-1/2007 when compared to A/SWG/
Nigeria/5388-8/2007 and A/SWG/Nigeria/5388-2/2007, A/
SWG/Nigeria/5388-5/2007, respectively. NA sequence variabil-
ity within the Nigerian viral genomes was less pronounced
(homology 99.0%­100%). However, 5 amino acid differences
were detected in A/SWG/Nigeria/5388-2-8/2007 when com-
pared to A/WFWD/Nigeria/3927-1/2007.
Satellite tracking
Movements of one of the H5N2 infected birds were tracked for
47 days, revealing that this bird survived infection by an HP viral
genotype. This adult male WFWD had been equipped with a
satellite transmitter on the date of capture and sampling, without
knowing its infection status. No clinical signs were reported for this
bird and it had a medium body mass respective to this species
(630g [14]). It first remained for 17 days within an 8 km radius of
the capture site, performing daily movement of 1­7 km (Figure 1).
It then flew eastwards to Chad in four stages (140­175 km daily
flights), covering a total distance of 655 km. The satellite signal
disappeared a few days after this duck had settled in the interior of
Lake Chad, at 360 km from the sampling site. We consider a
mortality associated with HPAI infection unlikely, since the time
span between capture and signal disappearance (47 days) is long,
and comparatively 6 times greater than the maximum time of
death which has been reported in susceptible wild waterfowl
following experimental infection with H5N1 HPAI viruses
(,8 days post inoculation; [15­18]). Signal loss, though specula-
tive, is likely related to the death of the bird (either natural or by
hunting) or to the failure of the battery of the transmitter. For
comparison, another WFWD fitted with a similar transmitter in
Nigeria, which tested negative for type A influenza virus (Table 1),
stopped transmitting after 25 days. During this period, this non-
infected WFWD (captured, sampled and equipped at the same site
on the same day) had a similar movement pattern to the H5N2-
infected WFWD, remaining within the same Hadejia-Nguru
Wetlands area (within 17 km radius of capture site during the
same first 18 day period), and performing similar short daily
movements (1­18 km).
Discussion
The H5N2 viruses we detected in two apparently healthy wild
waterfowl species constitute a rare finding of infection by AIVs
with an HPAI viral genotype in wild birds. Until recently, AIVs
with an HPAI viral genotype had not been detected in free-living
wild birds [5,6], except for a single report of an H5N3 HPAI virus
outbreak in South Africa [7], and a few cases in terrestrial birds
associated with outbreaks in poultry [6,19]. Since the emergence
and spread of the Asian H5N1 virus lineage over Eurasia and
Africa, H5N1 HPAI viruses have been isolated in a wide range of
wild bird species. However, both these H5N3 and H5N1 HPAI
viruses were isolated in sick, moribund or dead wild birds. Despite
extensive global wildlife surveillance efforts, no infection with
H5N1 HPAI viruses has been detected in healthy wild birds,
except for a few isolated cases [20,21]. Therefore, the significance
of wild birds as a source of infection and their influence on the
epidemiology of H5N1 HPAI viruses is yet to be fully established.
The two species in which these H5N2 viruses were detected are
both Anatidae species, a group of wild birds with a predominant
role in the perpetuation of LPAI viruses [2,22]. Both species forage
in shallow water and are highly gregarious outside the breeding
season, two ecological factors associated with an increased
exposure to AIV infection. The fact that we detected these H5N2
viruses at a three-day interval at the same site first suggested a virus
transmission between these two cohabiting species. WFWD and
SWG were the most abundant Afro-tropical ducks counted at the
sampling site (53% and 40% respectively, n = 3772). Moreover,
these two species, which commonly share the same habitats at
feeding and roosting sites and have similar foraging behaviour and
daily activity patterns [14], are prone to regular contacts or to
indirect AIV transmission via shared water. However, high
variation at the nucleotide and amino acid level was observed
between the Nigerian H5N2 HPAI sequences detected in the
WFWD and SWG. The presence of genetically distinguishable
H5N2 HPAI viruses does not support the hypothesis of a cluster of
infection. Rather, it indicates the co-circulation of distinct viruses in
the same waterfowl community, suggesting some degree of genetic
adaptation to different host species.
The survival and movements of one of the H5N2 infected duck
revealed by satellite telemetry provide evidence of a non-lethal
natural infection by an AIV with an HPAI viral genotype in a wild
duck. This result is consistent with the absence of disease signs or
mortality generally reported in ducks for most HPAI viruses.
Before the Asian H5N1 HPAI viruses, no disease or mortality
events associated with a natural AIV infection had been reported
in either wild or domestic ducks ([5]; with the single exception of
Table 2. Summary of the cleavage site motif in selected
African H5 HPAI and LPAI viruses.
Virus Cleavage site
No. basic
amino
acids
Nigerian H5N2 HPAI PQKEKRRKKR*GLF 7
PQREKRRKKR*GLF
African H5N1 HPAI 2006­2007a
PQGERRRKKR*GLF 6
A/chicken/Egypt/5610NAMRU3-F3/2006
(H5N1)
PQGKRRRKKR*GLF 7
A/chicken/Egypt/5611NAMRU3-AF/2006
(H5N1)
PQGKRRRKKR*GLF 7
A/Sudan/2006 (H5N1) HPAI PQGEGRRKKR*GLF 5
A/ostrich/South Africa/N227/04 (H5N2)
HPAI (9)
PQREKRRKKR*GLF 7
A/tern/South Africa/1961 (H5N3) HPAI PQRETRRQKR*GLF 5
A/ostrich/South Africa/AI1091/2006 (HPAI) PQRRKKR*GLF 5
A/ostrich/South Africa/AI1160/2006 (LPAI) PQRETR*GLF 2
A/yellow-billed duck/South Africa/811/04
(H5N1LPAI)
PQRETR*GLF 2
A/mallard/Bavaria/1/2005 (H5N2 LPAI) PQRETR*GLF 2
Cleavage site motif of the HA sequence phylogenetically most closely related to
the Nigerian H5N2 HA (A/mallard/Bavaria/1/2005) is included for comparison.
a
Representatives of all the sequenced African H5N1 HPAI viruses circulating in
2006­2007 (exceptions were observed in the Sudanese isolates and in 2
Egyptian isolates).
doi:10.1371/journal.ppat.1000127.t002
Highly Pathogenic Avian Influenza in Wild Birds
PLoS Pathogens | www.plospathogens.org 4 August 2008 | Volume 4 | Issue 8 | e1000127
Figure 2. Phylogenetic tree based on the sequence analysis of the entire segment encoding for HA proteins, with representative
H5N1, H5N2, and H5N3 influenza A viruses isolated in naturally infected wild and domestic birds in Asia, Europe, and Africa. Viral
sequences analyzed in this study are marked with a circle.
doi:10.1371/journal.ppat.1000127.g002
Highly Pathogenic Avian Influenza in Wild Birds
PLoS Pathogens | www.plospathogens.org 5 August 2008 | Volume 4 | Issue 8 | e1000127
Figure 3. Phylogenetic tree based on the sequence analysis of the entire segment encoding for NA proteins. The phylogenetic tree
includes selected N2 sequences of influenza viruses isolated in Asia, Europe and Africa. Viral sequences analyzed in this study are marked with a circle.
doi:10.1371/journal.ppat.1000127.g003
Highly Pathogenic Avian Influenza in Wild Birds
PLoS Pathogens | www.plospathogens.org 6 August 2008 | Volume 4 | Issue 8 | e1000127
mortality cases in domestic ducks and geese during H7N1 HPAI
outbreaks in Italy [23]). In addition, experimental infections with
most AIV with a high pathogenicity for gallinaceous poultry
(except H5N1), as well as with the A/tern/South Africa/1/61
(H5N3) [24], caused no clinical signs in domestic ducks [25] (except
for A/FPV/Rostock/34 (H7N1) [26]). Lethality of HPAI viruses for
duck has changed with the re-emergence of H5N1 HPAI viruses in
Hong Kong in 2002 [27­29]. These H5N1 HPAI viruses, previously
non lethal in ducks [30], have become highly lethal in some naturally
infected wild ducks [27,31], as well as in some experimentally
infected wild [15,18] and domestic ducks [28,32]. However, H5N1
HPAI viruses show a high diversity in their pathogenicity for ducks.
Results of experimental inoculations in these birds ranged from
asymptomatic infection to high mortality rate, depending on species
[15,18], virus strain and age of the host (i.e. lethal only in ,5-week-
old domestic ducks) [33].
Telemetry results further suggest that the AIV infection did not
reduce migration capacity of this duck. The movement pattern we
recorded is in agreement with the described movement behaviour of
WFWDs [14,34], which are mainly sedentary but perform
occasional flights of several hundred kilometres. However, the bird
we tracked performed some long distance movements only 18 days
after it had tested positive for HPAI virus. This period is longer than
the duration of illness and viral shedding generally recorded in
waterfowl inoculated with H5N1 HPAI viruses (,7 days; [15­
18,32]). Despite the absence of clinical symptoms at the time of
capture, we cannot conclude on the ability of a wild duck to fly great
distances during infection and hence to spread the virus, since no
data about the persistence of viral shedding could be collected.
However, the local movements (1­7 km) performed initially by this
infected bird were similar in pattern and range to the movements
concurrently monitored in one non-infected WFWD. Furthermore,
this pattern was comparable to the short weekly distances recorded
in two satellite tracked WFWD in South Africa [35], suggesting that
movements were not influenced by the AIV infection.
Based on the current OIE definition [36], the molecular
signature at the HA cleavage site, as revealed in the H5N2
Nigerian genomes, defines these viruses as HPAI. The acquisition
of multiple basic amino acid at the HA cleavage site is recognised
as a major molecular determinant of virulence in AIVs, but it
might not be sufficient for the expression of high lethality [13,37].
We did not succeed in isolating these viruses, thus in vivo
pathogenicity tests to establish the pathogenicity for chickens
[36] could not be performed. Based on the clear fluorescence
signal and the low Ct values in real time PCR, the failure of the
virus isolation attempt was probably not related to a sensitivity
issue of the virological method. Rather, the reasons for virus
isolation failure could be related to the loss of viral viability during
transportation and storage or to a lack of adaptation of the viruses
to the substrates used in the laboratory (i.e., SPF embryonating
chicken eggs) [38]. Caution should therefore be taken to consider
the phenotype of these Nigerian viruses as highly pathogenic for
chickens, since few exceptions have been described involving H5
viruses with a multiple basic amino acid motif resulting in low
pathogenicity in experimentally infected chickens [37,39]. How-
ever, the number of basic amino acids at the HA cleavage site of
the Nigerian H5N2 viruses was identical (n = 7) to some other
African H5N1 and H5N2 HPAI viruses (Table 2). Furthermore,
the cleavage site of A/SWG/Nigeria/5388-2-8-5/2007 is identical
to A/ostrich/South Africa/N227/2004 (H5N2) HPAI which
caused clinical signs characteristic of HPAI phenotype [11],
suggesting a potential for an HPAI behaviour of these viruses.
Despite the relative proximity of our sampling site with past and
recent H5N1 HPAI virus outbreaks in domestic birds in northern
Nigeria (#80 km), the H5N2 viruses we detected were unrelated
to any 2004-07 H5N1 HPAI strains from Asia, Europe or Africa.
The HA gene instead clustered with sequences from ducks from
South and Central Europe, in particular with contemporary
strains isolated from wild ducks, suggesting a connection via
migratory flyways. This finding is consistent with the close
relationship previously reported between H5N1 HPAI viruses
isolated from Nigerian poultry and European wild birds [40] and
with the role suggested for the Eurasian migratory birds in the
occurrence of H5 outbreaks previously reported in Southern
Africa [41,42]. The two waterfowl species we found infected in
Nigeria (i.e. WFWD and SWG) are widespread across sub-
Saharan Africa, but no African populations of either species move
out of the continent [14]. H5N2 viruses were detected during the
mid-dry season of the Sahel zone, when large flocks of Afro-
tropical duck species congregate at permanent water bodies in
large floodplains. At this time of year, these birds also mix with a
large number of migratory ducks originating from Eurasian
breeding grounds (e.g. 50,000 Garganeys Anas querquedula were
present on the sampling site), which gather in tropical wetlands
during the northern winter. This suggests a potential role of
Eurasian migratory ducks in the introduction of these viruses or
their precursors into these ecosystems, which offer an interface
between Eurasian and Afro-tropical wild ducks.
The proposed mechanism for the emergence of pathogenicity is
that it occurs only after a virus has been introduced and adapted to
gallinaceous poultry populations [5]. This theory is supported by
phylogenetic analysis that demonstrated shared phylogenetic sub-
lineages among viruses of both high and low pathogenicity [3,43],
and that identified precursors of HPAI viruses of gallinaceous poultry
in wild ducks [44,45]. However, the Nigerian HA sequences showed
little or no homology with recent H5N2 or H5N1 HPAI viruses
isolated from gallinaceous poultry in Europe and Africa. Further, no
mutations commonly observed in gallinaceous poultry-adapted AI
viruses were observed in their genomes [12,13,43], suggesting that
these viruses may not have circulated in gallinaceous poultry, at least
not in an extensive manner. This appears to be supported by the lack
of evidence of a circulation in poultry of H5 viruses different from
H5N1 HPAI during an extensive active surveillance conducted over
Nigeria in 2007 (T. Joannis, personal communication), or in samples
collected from suspected cases in Nigerian poultry in the same period
H5N2 was detected in wild waterfowl (January­February 2007) [46].
The absence of molecular features associated with an extensive viral
circulation in gallinaceous birds, together with the absence of
evidence of circulation of H5 LPAI precursor or HPAI viruses in
gallinaceous birds in Nigeria, suggests that the acquisition of an HP
viral genotype could have taken place in ducks, either wild or
domestic, or in other non-gallinaceous species.
The site where the birds were sampled constitutes one of the
major wetlands for waterbirds in Nigeria, but is also located in one
of the major areas of free-ranging duck production in this country
[47]. WFWD and SWG are found in all types of natural
freshwater habitats, but they also commonly feed on rice fields
where domestic ducks may also feed. Local conditions are hence
favourable to potential transmission events between wild and
domestic ducks, through direct contact or shared water.
Though several factors contribute to virulence in AIVs, the
acquisition of multiple basic amino acids at the HA cleavage site is
recognised as one of the major molecular determinants in the
development of HP strains [37]. The presence of coexisting but
genetically distinguishable avian influenza viruses with an HP viral
genotype in two cohabiting species of wild waterfowl, with evidence
of non-lethal infection at least in one species and without evidence of
prior extensive circulation of the virus in domestic poultry, suggest
Highly Pathogenic Avian Influenza in Wild Birds
PLoS Pathogens | www.plospathogens.org 7 August 2008 | Volume 4 | Issue 8 | e1000127
that some strains with a potential high pathogenicity for poultry
could be maintained in a community of wild waterfowl.
Methods
Wild bird surveillance
Birds were captured using mist nets with playbacks, baited traps
and nooses. Both cloacal and oropharyngeal swabs were collected
from live-caught birds using sterile cotton swabs (5615 mm and
2610 mm size, Dutscher manufacturer). Fresh faecal samples
were also collected at roosting areas (non-captured birds), ensuring
that the exact species yielding the samples were identified. The
health of birds was assessed by observing their movements prior to
sampling, and for captured birds, by their behaviour in the hand,
mass, and body condition. Samples were placed in a transport
medium consisting of an isotonic phosphate buffered saline (PBS),
pH 7.0­7.4, containing antibiotics and antimycotic (penicillin
10,000 units/ml, streptomycin 10 mg/ml, nystatin 1000 U/ml
and gentamycin 250 mg/ml) supplemented with 20% glycerol.
Samples were stored in liquid nitrogen directly in the field. An
unbroken cold chain was maintained, using a liquid nitrogen
container during national transport and a cryopack with dry ice
during international shipment to the laboratory.
RNA isolation, amplification, and sequence analysis
RNA was extracted at the CIRAD laboratory (Montpellier,
France) with the Nucleospin RNA virus kit from Macherey Nagel
using the automate Biomek FX from Beckman, after elution into
50 ml H2O. RNA was first screened for the presence of genomic
nucleic acid from type A Influenza viruses by means of RRT-PCR
targeting the influenza matrix gene [48]. Positive samples were tested
by RT-PCR specific for H5 and H7 subtype [48,49]. Conventional
RT-PCR amplifying specifically the cleavage site was carried out on
H5 or H7 positive samples and amplified products sequenced in
order to detect the presence of multibasic amino acids. At the OIE/
FAO Reference Laboratory for AI (IZSE, Padova, Italy), confirma-
tory molecular tests, including sequencing, molecular and phyloge-
netic analysis, were further applied and virus isolation in
embryonating SPF chicken eggs was attempted for all RRT-PCR-
positive samples according to standard procedures [36].
From positive samples the complete ORF of the HA and NA
genomic segments were directly sequenced and phylogenetically
analysed. Briefly, the cDNA of viral genomic segments 4 and 6 was
amplified through the application of five to six distinct PCRs
targeting overlapping regions of approximately 400 to 900 bp
each. Briefly, 6 primer sets were applied to amplify and sequence 6
overlapping segments for the HA gene sequence (59-39 segments
position 222 to 666; 422 to 1303; 900 to 1726; 557 to 1526; 729
to 1207 and 437 to 1248; referring to A/turkey/Italy/1325/
05(H5N2); Genbank accession number CY022629). For the NA
gene, 5 primer sets were applied to amplify and sequence five
overlapping segments (59-39 segments position 27 to 171; 1­723;
583 to 950; 877 to 1374 and 877 to 1426; referring to A/turkey/
Italy/1325/05(H5N2); Genbank accession number CY022631).
Primer sequences are available on request.
Amplicons were purified (ExoSap-IT, USB Corporation, Ohio,
USA) and sequenced on both strands. The obtained sequences
were then aligned together with contemporary and Eurasian
HPAI and LPAI sequences and their phylogenetic relationship was
inferred by the application of Neighbor-Joining algorithm (MEGA
4.0; 1,000 bootstrapping) [50]. Sequences have been deposited in a
public database (GenBank accession numbers EU544242 to
EU544248).
Satellite tracking
The two WFWDs captured at the Dagona Waterfowl sanctuary
were equipped with an 18g solar powered satellite Platform
Transmitter Terminal (PTT), attached on the back of the bird using
a Teflon harness-attachment. Sex was determined on the basis of
cloacal examination, and body mass was measured to the nearest
10 g. Before the deployment of transmitters in Nigeria a test had
been conducted on captive WFWDs at Montpellier Zoo (France) to
monitor the effects of attachment technique and transmitter load on
birds. Movements were monitored using the Argos satellite tracking
system. The PTT was programmed to transmit for a 10 h interval
every 24 h according to the battery capacity. Argos CLS (Toulouse,
France) processed the satellite signals and provided locations. Only
locations with a precision ,1000 m (location classes 1, 2 and 3,
dependent on the number and quality of signals received) were used
for analysis and mapping.
Acknowledgments
We are grateful to all the ornithologists and veterinarians who collaborated
in this large-scale surveillance programme coordinated by CIRAD and
Wetlands International, and in particular to the participants of the
Nigerian field operation: Onyeche O. Adah, Delmo Mannok, Longtong
Turshat (AP Leventis Ornithological Research Institute; contribution
number 31), Harry Hanson Jr (Hadejia-Nguru Wetlands Office), Giuseppe
Rossi (Istituto Nazionale per la Fauna Selvatica), Dr Scott Petrie (Bird
Studies Canada) and Gabriel Norevik, Per O
¨ sterman, and Andreas
Eriksson (Ottenby Bird Observatory; contribution number 225).
The authors thank the Conservator-General of Nigerian National Parks
for permission to capture birds for telemetry study in Nigeria, and to Celia
Abolnik (Ondesterpoort Veterinary Institute of Pretoria, South Africa) for
providing the information on the South African AI sequences included in
this study. We would like to thank the Lunaret-Montpellier Zoo (France)
for allowing the test on captive birds to be conducted.
Author Contributions
Conceived and designed the experiments: NG SHN WH TD JL JD FM.
Performed the experiments: SH WH JYT TJ PG SM PM. Analyzed the
data: NG GC JC IM AF. Contributed reagents/materials/analysis tools:
GC SH UO JHM. Wrote the paper: NG GC. Revised the article critically
for important intellectual content: IC.
References
1. Webster RG, Bean WJ, Gorman OT, Chambers TM, Kawaoka Y (1992)
Evolution and ecology of influenza A viruses. Microbiol Rev 56: 152­179.
2. Olsen B, Munster VJ, Wallensten A, Waldenstrom J, Osterhaus AD, Fouchier RA
(2006) Global patterns of influenza a virus in wild birds. Science. 312: 384­88.
3. Ro
¨hm C, Horimoto T, Kawaoka Y, Suss J, Webster RG (1995) Do
hemagglutinin genes of highly pathogenic avian influenza viruses constitute
unique phylogenetic lineages? Virology 209: 664­670.
4. van Gils JA, Munster VJ, Radersma R, Liefhebber D, Fouchier RAM, et al.
(2007) Hampered Foraging and Migratory Performance in Swans Infected with
Low-Pathogenic Avian Influenza A Virus. PLoS ONE 2(1): e184.
5. Alexander DJ (2000) A review of avian influenza in different bird species. Vet
Microbiol 74: 3­13.
6. Stallknecht DE, Nagy E, Hunter D, Slemons RD (2007) Avian Influenza. In:
Thomas NJ, Hunter DB, Atkinson CT, eds. Infectious Diseases of Wild Birds
Blackwell Publishing. pp 108­130.
7. Becker WB (1966) The isolation and classification of tern virus: influenza virus
A/tern/South Africa/1961. J Hyg 64: 309­320.
8. Food and Agriculture Organization of the United Nations (2007) Wild Birds and
Avian Influenza: an introduction to applied field research and disease sampling
techniques. Whitworth D, Newman SH, Mundkur T, Harris P, eds. FAO
Animal Production and Health Manual, No. 5, pp 123 (available at www.fao.
org/avianflu).
9. Gaidet N, Dodman T, Caron A, Balanc
¸a G, Desvaux S, et al. (2007) Avian
influenza viruses in water birds, Africa. Emerg Infect Dis 13: 626­629.
Highly Pathogenic Avian Influenza in Wild Birds
PLoS Pathogens | www.plospathogens.org 8 August 2008 | Volume 4 | Issue 8 | e1000127
10. De Benedictis P, Joannis TM, Lombin LH, Shittu I, Beato MS, Rebonato V,
Cattoli G, Capua I (2007) Field and laboratory findings of the first incursion of
the Asian H5N1 highly pathogenic avian influenza virus in Africa. Avian Pathol
36: 115­117.
11. Abolnik C (2007) Molecular characterization of H5N2 avian influenza viruses
isolated from South African ostriches in 2006. Avian Dis 51: 873­879.
12. Matrosovich M, Zhou N, Kawaoka Y, Webster RG (1999) The surface
glycoproteins of H5 influenza viruses isolated from humans, chickens, and wild
aquatic birds have distinguishable properties. J Virol 73: 1146­1155.
13. Banks J, Speidel ES, Moore E, Plowright L, Piccirillo A, et al. (2001) Changes in
the haemagglutinin and the neuraminidase genes prior to the emergence of
highly pathogenic H7N1 avian influenza viruses in Italy. Arch Virol 146:
963­973.
14. Brown LH, Urban EK, Newman K (1982) The Birds of Africa Vol. 1. London:
Academic Press. 521 p.
15. Brown JD, Stallknecht DE, Beck JR, Suarez DL, Swayne DE (2006)
Susceptibility of North American ducks and gulls to H5N1 highly pathogenic
avian influenza viruses. Emerg Infect Dis 12: 1663­1670.
16. Pasick J, Berhane Y, Embury-Hyatt C, Copps J, Kehler H, et al. (2007)
Susceptibility of Canada geese (Branta canadensis) to highly pathogenic avian
influenza virus (H5N1). Emerg Infect Dis 13: 1821­1827.
17. Brown JD, Stallknecht DE, Swayne DE (2008) Experimental infection of swans
and geese with highly pathogenic avian influenza virus (H5N1) of Asian lineage.
Emerg Infect Dis 14: 136­142.
18. Keawcharoen J, van Riel D, van Amerongen G, Bestebroer T, Beyer WE, et al.
(2008) Wild ducks as long-distance vectors of highly pathogenic avian influenza
virus (H5N1). Emerg Infect Dis 14: 600­607.
19. Capua I, Grossele B, Bertoli E, Cordioli P (2000) Monitoring for highly
pathogenic avian influenza in wild birds in Italy. Vet Rec 147: 640.
20. Saad MD, Ahmed LS, Gamal-Eldein MA, Fouda MK, Khalil FM, et al. (2007)
Possible avian influenza (H5N1) from migratory bird, Egypt. Emerg Infect Dis
13: 1120­1121.
21. Chen H, Li KS, Wang J, Fan XH, Rayner JM, et al. (2006) Establishment of
multiple sublineages of H5N1 influenza virus in Asia: Implications for pandemic
control. Proc Nat Acad Sci USA 103: 2845­50.
22. Wallensten A, Munster VJ, Latorre-Margalef N, Brytting M, Elmberg J, et al.
(2007) Surveillance of influenza A virus in migratory waterfowl in Northern
Europe. Emerg. Infect. Dis. 13: 404­411.
23. Capua I, Mutinelli F (2001) Mortality in Muscovy ducks (Cairina moschata) and
domestic geese (Anser anser var. domestica) associated with natural infection with a
highly pathogenic avian influenza virus of H7N1 subtype. Avian Pathol 30:
179­183.
24. Kishida N, Sakoda Y, Isoda N, Matsuda K, Eto M, Sunaga Y, et al. (2005)
Pathogenicity of H5 influenza viruses for ducks. Arch Virol 150: 1383­1392.
25. Wood GW, Parsons G, Alexander DJ (1995) Replication of influenza A viruses
of high and low pathogenicity for chickens at different sites in chickens and ducks
following intranasal inoculation. Avian Pathol 24: 545­541.
26. Alexander DJ, Allan WH, Parsons DG, Parsons G (1978) The pathogenicity of
four avian influenza viruses for fowls, turkeys and ducks. Res Vet Sci 24:
242­247.
27. Ellis TM, Bousfield B, Bissett L, Dyrting K, Luk GSM, et al. (2004) Investigation
of outbreaks of highly pathogenic H5N1 avian influenza in waterfowl and wild
birds in Hong Kong in late 2002. Avian Pathol 33: 492­505.
28. Sturm-Ramirez KM, Ellis T, Bousfield B, Bissett L, Dyrting K, et al. (2004)
Reemerging H5N1 influenza viruses in Hong Kong in 2002 are highly
pathogenic to ducks. J Virol 78: 4892­4901.
29. Webster RG, Hulse-Post DJ, Sturm-Ramirez KM, Guan Y, Peiris M, et al.
(2007) Changing epidemiology and ecology of highly pathogenic avian H5N1
influenza viruses. Avian Dis 51: 269­272.
30. Perkins LEL, Swayne DE (2002) Pathogenicity of a Hong Kong-origin H5N1
highly pathogenic avian influenza virus for emus, geese, ducks, and pigeons.
Avian Dis 46: 53­63.
31. Pittman M, Laddomada A, Freigofas R, Piazza V, Brouw A, et al. (2007)
Surveillance, prevention, and disease management of avian influenza in the
European Union. J Wildlife Dis 43: S64­S70.
32. Hulse-Post DJ, Sturm-Ramirez KM, Humberd J, Seiler P, Govorkova EA, et al.
(2005) Role of domestic ducks in the propagation and biological evolution of
highly pathogenic H5N1 influenza viruses in Asia. Proc Natl Acad Sci U S A
102: 10682­10687.
33. Pantin-Jackwood MJ, Suarez DL, Spackman E, Swayne DE (2007) Age at
infection affects the pathogenicity of Asian highly pathogenic avian influenza
H5NI viruses in ducks. Virus Res 130: 151­161.
34. Oatley TB, Prys-Jones RP (1986) A comparative analysis of movements of
southern African waterfowl (Anatidae) based on ringing recoveries. S Afr J Wildl
Res 16: 1­6.
35. Petrie SA, Rogers KH (1997) Satellite tracking of white-faced whistling ducks in
a semiarid region of South Africa. J Wildlife Manage 61: 1208­1213.
36. OIE (World Organization for Animal Health) (2004) Highly pathogenic avian
influenza. In: Manual of diagnostic tests and vaccines for terrestrial animals, 5th
Edition, Paris, France: Office International des Epizooties. 258 p.
37. Lee CW, Lee YJ, Swayne D, Senne D, Linares DJ, et al. (2007) Assessing
potential pathogenicity of avian influenza virus: Current and experimental
system. Avian Dis 51: 260­263.
38. Fouchier RAM, Olsen B, Bestebroer TM, Herfst S, van der Kemp L, et al.
(2003) Influenza A virus surveillance in wild birds in northern Europe in 1999
and 2000. Avian Dis 47: 857­860.
39. Lo
¨ndt BZ, Banks J, Alexander DJ (2007) Highly pathogenic avian influenza
viruses with low virulence for chickens in in vivo tests. Av Pathol 36: 347­350.
40. Ducatez MF, Olinger CM, Owoade AA, De Landtsheer S, Ammerlaan W, et al.
(2006) Avian flu: Multiple introductions of H5N1 in Nigeria. Nature 442: 37.
41. Abolnik C, Cornelius E, Bisshopp SPR, Romito M, Verwoerd DJ (2006)
Phylogenetic analyses of genes from South African LPAI viruses isolated in 2004
from wild aquatic birds suggests introduction by Eurasian migrants. Dev. Biol.
24: 189­199.
42. Olivier AJ (2006) Ecology and epidemiology of avian influenza in ostriches. Dev
Biol 124: 51­57.
43. Banks J, Speidel EC, McCauley JW, Alexander DJ (2000) Phylogenetic analysis
of H7 haemagglutinin subtype influenza A viruses. Arch Virol 145: 1047­1058.
44. Munster VJ, Wallensten A, Baas C, Rimmelzwaan GF, Schutten M, et al. (2005)
Mallards and highly pathogenic avian influenza ancestral viruses, Northern
Europe. Emerg Infect Dis 11: 1545­1551.
45. Campitelli L, Di Martino A, Spagnolo D, Smith GJD, Di Trani L, et al. (2008)
Molecular analysis of avian H7 influenza viruses circulating in Eurasia in 1999-
2005: detection of multiple reassortant virus genotypes. J Gen Virol 89: 48­59.
46. Monne I, Joannis TM, Fusaro A, de Benedictis P, Lombin LH, et al. (2008)
Reassortant avian influenza virus (H5N1) in poultry, Nigeria, 2007. Emerg
Infect Dis 14: 637­640.
47. Adene DF, Oguntade AE (2006) The structure and importance of the
commercial and village based poultry industry in Nigeria. Rome: FAO. 109 p.
48. Spackman E, Senne DA, Myers TJ, Bulaga LL, Garber LP, et al. (2002)
Development of a real-time reverse transcriptase PCR assay for Type A
influenza virus and the avian H5 and H7 hemagglutinin subtypes. J Clin
Microbiol 40: 3256­3260.
49. Slomka MJ, Pavlidis T, Banks J, Shell W, McNally A, et al. (2007) Validated H5
Eurasian realtime reverse transcriptase­polymerase chain reaction and its
application in H5N1 outbreaks in 2005­2006. Avian Dis 51: 373­377.
50. Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: Molecular Evolutionary
Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol 24: 1596­1599.
Highly Pathogenic Avian Influenza in Wild Birds
PLoS Pathogens | www.plospathogens.org 9 August 2008 | Volume 4 | Issue 8 | e1000127
