In vitro characterization of naturally occurring influenza H3NA− viruses lacking the
NA gene segment: Toward a new mechanism of viral resistance?
V. Moulesa,⁎,1, O. Ferrarisa,1, O. Terriera, E. Giudiceb, M. Yvera, J.P. Rollandb, M. Bouscambert-Duchampa,c,
C. Bergerona, M. Ottmanna, E. Fournierd, A. Traversiera, C. Boulee, A. Rivoiree, Y. Linf, A. Hayf, M. Valettea,c,
R. Marquetd, M. Rosa-Calatravaa, N. Naffakhg, G. Schoehnh,i, D. Thomasb, B. Linaa,c
aUniversité de Lyon, F-69000, Lyon, France; Université Lyon 1, Faculté de médecine RTH Laennec; CNRS FRE 3011 VirPath, Virologie et Pathologie Humaine, F-69008, Lyon, France
bUniversité Rennes 1, CNRS Interactions Cellulaires et Moléculaires, UMR 6026, Campus de Beaulieu, bâtiment 13, F-35042 Rennes cedex, France
cLaboratoire de virologie, centre de Biologie et de Pathologie Est, Centre National de référence influenza région sud, WHO collaborating centre for influenza,
Hospices Civils de Lyon, 59, boulevard Pinel, 69677 Bron cedex, France
dUPR9002 du CNRS, Institut de Biologie Moléculaire et Cellulaire, 15 rue René Descartes, Strasbourg 67084 cedex, France
eCentre Technologique des Microstructures, EZUS LYON 1, UCBL, Villeurbanne, France
fMRC-National Institute for Medical Research and WHO reference center for influenza, Mill Hill, London NW7 1AA, UK
gInstitut Pasteur, Unité de Génétique Moléculaire des Virus à ARN, URA 3015 CNRS, Université Paris Diderot-Paris 7, 25 rue du Dr Roux 75015 Paris, France
hIBS, UMR 5075 CEA-CNRS-UJF. 41 rue Jules Horowitz 38 027 Grenoble Cedex 1, France
iUVHCI, UMI 3265 UJF-EMBL-CNRS, 6, rue Jules Horowitz, 38042 Grenoble, France
a b s t r a c ta r t i c l e i n f o
Received 2 March 2010
Returned to author for revision
30 March 2010
Accepted 27 April 2010
Available online 3 June 2010
Among a panel of 788 clinical influenza H3N2 isolates, two isolates were characterized by an oseltamivir-
resistant phenotype linked to the absence of any detectable NA activity. Here, we established that the two
H3NA− isolates lack any detectable full-length NA segment, and one of these could be rescued by reverse
genetics in the absence of any NA segment sequence. We found that the absence of NA segment induced a
moderate growth defect of the H3NA− viruses as on cultured cells. The glycoproteins density at the surface
of H3NA− virions was unchanged as compared to H3N2 virions. The HA protein as well as residues 188 and
617 of the PB1 protein were shown to be strong determinants of the ability of H3NA− viruses to grow in the
absence of the NA segment. The significance of these findings about naturally occurring seven-segment
influenza A viruses is discussed.
© 2010 Elsevier Inc. All rights reserved.
Human influenza viruses are responsible for mild to severe res-
piratory tract infections. Influenza A viruses present two major sur-
face glycoproteins, the hemagglutinin (HA) and the neuraminidase
(NA). The HA is a trimeric type 1 membrane glycoprotein involved in
the binding to cell surface sialoconjugate receptors and in the fusion
between the viral envelope and cellular endosomal membranes. The
NA removes sialic acids from the HA protein as well as from cellular
glycolipids or glycoproteins, thereby preventing the aggregation
of virus particles and allowing the release of virus from host cell
receptors(Palese etal., 1974). Thesialidase activityof theNA may also
enable the virus to diffuse through the abundant mucin layer at the
surface of the respiratory epithelium. An optimal balance between
the sialic-acid binding activity of the HA and the sialidase activity of
the NA appears critical for viral fitness (Mitnaul et al., 2000; Wagner
et al., 2002).
The control of human seasonal influenza can be achieved by vac-
pandemic, as illustrated by the current swine-origin H1N1 pandemic,
vaccines are generally not available during the first phase and the
use of antivirals is considered the best treatment option during this
initial period. Since 1999, neuraminidase inhibitors (NAIs), zanam-
ivir and oseltamivir, have been used in the treatment of seasonal
influenza. The rate of emergence of resistant isolates after NAI
treatment has been found to range from 0.4% in adults to up to 4%
in children (Whitley et al., 2001; Roberts, 2001). Substitutions on
catalytic or framework residues of the NA active site are responsible
for the cases of resistance observed in vivo on NAI-resistant viruses
recovered from both drug-treated and untreated patients as well as
in vitro on viruses recovered from culture experiments in the presence
of NAI selective pressure (for a review, see Ferraris and Lina, 2008).
Virology 404 (2010) 215–224
⁎ Corresponding author. Fax: +33 478778751.
E-mail address: firstname.lastname@example.org (V. Moules).
1VM and OF contributed equally to this work.
0042-6822/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/yviro
The implication of NA on replication ability is observed on variants
with mutated catalytic residues which showed reduced rates of repli-
naturally resistant to oseltamivir that emerged in 2007–2008 (http://
en/index.html) (Dharan et al., 2009). In contrast, variants with mu-
tations on framework residues retained an efficient transmissibility
(Gubareva et al., 1998). Moreover, selective pressure in vitro allowed
the recovery of H1N1 viruses with a truncated NA gene segment
lacking the region encoding the NA active site (Nedyalkova et al.,
2002). Such variants with deletions were found to grow on cultured
cells only in the presence of exogenous NA (Hughes et al., 2000).
In this case, HA compensatory mutations are not always observed
(Hughes et al., 2000; Gubareva et al., 2001). However, substitution on
HA gene that may lead to a reduce MDCK receptors binding and a
decreased dependence on NA activity has already been observed and
Abed et al., 2002).
In a previous study, we analyzed the susceptibility to NAIs of
human influenza A and B viruses isolated in patients presenting with
acute respiratory infection between 2002 and 2005 (Ferraris et al.,
2006). Overall, 788 viruses were tested for their NA activity using
a fluorometric assay. We detected two H3 viruses that did not show
any detectable NA activity even in the absence of NAI the A/Lyon-
CHU/26430/03 and A/Reunion/586/04 isolates, antigenically related
to A/Wyoming/3/03 and A/Wellington/1/04, respectively (Ferraris
et al., 2006). Attempts to amplify the NA gene and to detect the NA
protein by Western blot on the initial clinical samples as well as on
amplified viruses both failed (Ferraris et al., 2006).
In the present study, we further characterized the two H3NA− A/
Lyon-CHU/26430/03 and A/Reunion/586/04 isolates. We estab-
lished that they lack any detectable full-length or partial-length NA
segment, and that a recombinant A/Reunion/586/04 virus can be
produced by reverse genetics in the absence of the NA segment. We
examined the impact of the absence of NA segment and NA protein on
the growth properties and the morphology of the H3NA− viruses as
compared to H3N2 counterparts, and we investigated the molecular
basis of their ability to grow in the absence of the NA segment.
Growth kinetics of H3NA− viruses on cultured cells
Two influenza viruses, A/Lyon-CHU/26430/03 and A/Reunion/
586/04, showed no detectable sialidase activity when tested out of
788 A(H3N2) viruses isolated during the 2002–2005 period of time
could be detected after extraction of viral RNA and RT-PCR ampli-
fication using both primers specific for the untranslated regions of
the N2 segment (Hoffmann et al., 2000) and internal N2 primer sets
(data not shown). To further characterize these two H3NA− viruses,
plaque purification was performed and 30 plaque-purified viruses
were analyzed. The absence of any detectable NA activity and any
detectable full-length or partial-length NA segment was confirmed
on all purified viruses (data not shown), strongly suggesting that
the genome of the A/Lyon-CHU/26430/03 and A/Reunion/586/04
viruses consists of only seven genomic segments.
Upon amplification on MDCK cells, the H3NA− viruses showed
properties similar to contemporary H3N2 viruses, i.e. the ability to
agglutinate guinea pig but not chicken erythrocytes, a mild cytopathic
bacterial sialidase (data not shown). To compare the growth kinetics of
H3N2 and H3NA− viruses, MDCK cells were inoculated at a multiplicity
of infection (MOI) of 10−4and the release of viral progeny into the
supernatant was monitored up to 72 h post infection (p.i.) by
determining the infectious titers or by quantifying the amounts of M
genomic segment (M-vRNA) by real-time RT-PCR. As shown in Fig. 1A,
the infectious titers in the supernatant of MDCK cells infected with the
a maximal titer of 107.3TCID50/ml at 42 h pi. The growth curves of the
two H3NA− viruses A/Lyon-CHU/26430/03 and A/Reunion/586/04
appeared delayed. Titers measured at 24 h pi for the H3NA− viruses
were about 2 log lower compared to A/California/07/04, but maximal
titers measured at 66 h pi were in the same range (about 107.4TCID50/
Fig. 1. Kinetics of replication in vitro. A.MDCK cells were inoculated with influenza virus
at a MOI of 0,0001 in EMEM+1 µg/ml of trypsin, A/California/07/04 H3N2, A/Lyon-
CHU/26430/03 H3NA−, A/Reunion/586/04 H3NA−. After 1 h at 34 °C, the superna-
tant was discarded, the cells were washed with PBS 1× and fresh medium was added.
Samples of supernatants were harvested at predefined time points. The virus released
was then quantified by end point titration. B. MDCK cells were inoculated with
influenza virus at a MOI of 0.0001 in EMEM+1 µg/ml of trypsine, A/Moscow/10/99
H3N2, A/California/07/04 H3N2, A/Lyon-CHU/26430/03 H3NA−, A/Reunion/586/04
H3NA−. After 1 h at 34 °C, the supernatant was discarded, the cells were washed with
PBS 1× and fresh medium was added. Samples of supernatants were harvested at
predefined time points. The virus released was then quantified by M RNA quantification
by real time RTPCR. C. MDCK-SIAT 1 cells were inoculated with influenza virus at a MOI
of 0.0001 in EMEM+1 µg/ml of trypsine, A/Moscow/10/99 H3N2, A/California/07/04
H3N2, A/Lyon-CHU/26430/03 H3NA−, A/Reunion/586/04 H3NA−. After 1 h at 34 °C,
the supernatant was discarded, the cells were washed with PBS 1× and fresh medium
was added. Samples of supernatants were harvested at predefined time points. The
virus released was then quantified by M RNA quantification by real time RTPCR.
V. Moules et al. / Virology 404 (2010) 215–224
of the M-vRNA in the supernatant of MDCK infected cells showed the
same trend, although maximal concentrations were reached simulta-
neously at 42 h pi for the A/California/07/04 and for H3NA− viruses
curve very similar to the H3NA− viruses (Fig. 1B).
MDCK cells that stably over-express the human (alpha)2-6
sialyltransferase (MDCK-SIAT1 cells), proved to be a more suitable
system than MDCK cells to monitor the role of neuraminidase during
the growth process (Matrosovich et al., 2003) probably due to the fact
that the high amounts of sialyl-α2,6-Galactose-containing receptors
at the surface of MDCK-SIAT1 cells increase viral dependency on
NA sialidase activity. The kinetics of M-vRNA accumulation in the
supernatant of MDCK-SIAT cells infected with H3N2 and H3NA−
viruses were compared. The maximal concentrations of M vRNA
measured at 68 h pi were about two log higher for the H3N2 strains
(1011.7copies/ml) as compared to the H3NA− viruses (108.5and 109.2
copies/ml for the A/Lyon-CHU/26430/03 and A/Reunion/586/04
viruses, respectively) (Fig. 1C). Altogether, our data confirmed that
theH3NA− viruseswereable toreplicatein theabsenceof NA activity
on MDCK-SIAT1 cells, despite 2–3 logs growth defect.
The H3NA− cell growth abilities were also checked by seven
MDCK SIAT1 cell passages showing the viral stability of these viruses
(data not shown).
Morphological analysis of H3NA− virions by cryo-electron microscopy
To determine the impact of the lack of NA segment and NA protein
on viral morphology, ice-embedded virions from purified A/Califor-
nia/07/04 H3N2, A/Reunion/586/04 H3NA− and A/Lyon-CHU/
26430/03 H3NA− were observed by cryo-electron microscopy
(cryo-EM). Two independent viral preparations of each strain were
analyzed. Electron micrographs were recorded under low-dose
conditions at liquid nitrogen temperature with a Technai Sphera
LaB6 200 kV microscope. Images were collected at 30,000× magnifi-
cation in a defocus range between 2 and 3 µm. All observed particles
We decided against using the classification previously proposed by
and internal organization by cryo-electron tomography, since these
virions were classified by shape into three groups based on the axial
ratios. Viruses with an axial ratio ranging between 1.0 and 1.2, 1.2 and
1.6, or N1.6 were respectively classified as spherical, spheroidal or
elongated. The three classes of particle shapes were equally repre-
sented for the A/California/07/04 H3N2 viruses, with 37.9% (n=22)
spherical, 32.8% spheroidal (n=19) and 29.3% elongated (n=17)
virions (Fig. 2 and Table 1). In contrast, for the A/Reunion/586/04
H3NA− viruses, only spherical (50%, n=14) and elongated particles
and 30.0% (n=9) were spheroidal (Fig. 2 and Table 1). The spherical
A/California/07/04 particles had a reduced mean diameter (88.5±
9.3 nm) compared to the A/Reunion/586/04 and A/Lyon-CHU/
26430/03 particles (96.6±16.7 and 102.7±10.5 nm, respectively)
(Table 1) whereas no differences in diameter were observed among
spheroidal and elongated virions. These observations were in
shape and size of influenza A viruses (Booy et al., 1985;Fujiyoshi et al.,
1994; Harris et al., 2006).
Fig. 2. Ice-embedded H3N2 and H3NA− influenza virions observed in cryo-electron microscopy, produced and purified from MDCK cells. A. Representative micrographs of H3NA−
A/Reunion/586/04. B, C, D respectively zoom vignettes of H3N2 A/California/10/04, H3NA− A/Reunion/586/04, and A/Lyon-CHU/26430/03. Scale bars 100 nm. Arrows in D
indicate the glycoprotein spikes.
V. Moules et al. / Virology 404 (2010) 215–224
The stalks of the spike glycoproteins (GPs) were easily distin-
guishable (Fig. 2D, arrow) between the lipid bilayer and the electron
opaque layer formed by the GP globular heads, allowing us to count
theGPsandtomeasuretheaverage spacingfor agroupof4 GPs(4 GP-
spacing). The average spacing measured for the A/California/07/04
H3N2 virus was 26.9±8.8 nm (n=100, Table 1). From this data
we could calculate the average surface corresponding to 4×4 GPs,
and an estimated number of 695 surface GPs for a spherical virion
of 100 nm in diameter. Analysis of the A/Moscow/10/99 H3N2 strain
led to similar findings (data not shown). The average 4 GP-spacing
− viruses was 27.0±1.7 nm (n=80) and 28.1±1.4 nm (n=48),
respectively, which corresponded to estimated numbers of 689 and
637 surface GPs per spherical virion of 100 nm in diameter, respectively
(Table 1). The average 4 GP-spacing values showed no significant
differences between the different shape classes for a given virion
(Table 1), and between the A/California/07/04, A/Reunion/586/04
and A/Lyon-CHU/26430/03 viruses (pN0.01). Altogether, our data
suggest that despite the lack of NA protein, H3NA− viruses display
similar surface GP densities as compared to H3N2 viruses.
Ultrastructural analysis of H3NA− virions by electron microscopy
microscopy ultrathin sections of A/Reunion/586/04, A/Lyon-CHU/
24 h following low-multiplicity infection were compared. Represen-
tative results are shown in Fig. 3. Transversal sections of budding
virions appeared spherical in shape, with a diameter of 90–100 nm,
whereas longitudinal sections virion appeared elongated and pre-
sented different lengths (data not shown), in agreement with our
cryo-EMdata.The majorityof A/California/07/04transversalsections
showed the same electron dense-dot organization, i.e. a central dot
surrounded by seven other dots (Fig. 3A). Such an organization, each
dot likely representing a viral ribonucleoprotein, has previously
described by Noda et al. (2006). In contrast, a significant proportion
of A/Reunion/586/04 and A/Lyon-CHU/26430/03 transversally sec-
24% (n=410) and 27% (n=466) for A/Reunion/586/04 and A/Lyon-
CHU/26430/03, respectively, compared to 6% (n=453) for A/
California/07/04 (Fig. 3B and C). These observations were in
agreement with the absence of detectable NA segment in the A/
Reunion/586/04 and A/Lyon-CHU/26430/03 viral stocks, and with
the model for selective packaging, according to which packaging of a
full set of genomic segments depends on cis-acting signals present on
each of the eight genomic segments (Fujii et al., 2003).
Reverse genetics analysis of the H3NA− viral phenotype
To confirm the capacity of the A/Reunion/586/04 virus to
replicate in the absence of the NA segment, the cDNAs corresponding
to the seven remaining genomic segments were cloned in the
bidirectional reverse genetics (RG) plasmid pHW2000 (Hoffmann et
al., 2000). A set of eight pHW2000-derived RG plasmids was
constructed for the A/Moscow/10/99 H3N2 strain (MO). The seven
A/Reunion/586/04-derived RG plasmids, either or not supplemented
with the A/Moscow/10/99-derived NA plasmid, were co-transfected
into 293T cells. Co-transfection of eight (whole sets) or seven (whole
sets minus the NA plasmid) A/Moscow/10/99-derived RG plasmids
was used as a control. Transfection supernants were harvested at 72 h
post-transfection and diluted 1/10 to infect confluent monolayers of
MDCK cells. The viral titers and the amounts of M-vRNA in MDCK cells
supernatants were evaluated at 72 h post-infection by real-time RT-
PCR. About 1011copies of M-vRNA/ml were detected upon transfec-
tion of the eight A/Moscow/10/99-derived RG plasmids (Fig. 4,
MO, hatched bar). In contrast, in the absence of the NA plasmid, no
hemagglutinating activity and M-vRNA concentrations lower than 105
copies/ml were detected (Fig. 4, MO, gray bar). Similar results were
obtained with the A/Wellington/1/04 H3N2 virus (data not shown).
The low M-vRNA levels detected in the absence of NA most likely
corresponded to the detection of residual pHW2000-M plasmid. In
contrast, titers 1011copies of M-vRNA/ml were detected upon trans-
fection of the seven A/Reunion/586/04-derived RG plasmids (Fig. 4,
R2, gray bar), and the addition of the A/Moscow/10/99-derived NA
Image analysis of viral particles in cryo-EM. Shapes, diameters, axis lengths and determination of GP spike number at the viral surface.
Viruses Shapes, diameters and axis lengths (average±SD, nm)GP spikes
4 GP spacing (average±SD, nm)GP/ 100 nm spherical virion
26.9±8.8 (n=100) 695
27.0±1.7 (n=80) 689
Fig. 3. Ultrathin section electron microscopy micrographs of budding influenza H3N2 and H3NA− viruses. A. A/California/07/04, B. A/Reunion/586/04, C. A/Lyon-CHU/26430/03.
Black scale bar: 200 nm.
V. Moules et al. / Virology 404 (2010) 215–224
plasmid did not induce significant changes (Fig. 4, R2, hatched bar).
The viral stability of the recombinant R2 viruses has been controlled
after three MDCK cells passages (data not shown).
These data suggested that some genetic features of the A/
Reunion/586/04 genome accounted for its unique ability to grow
efficiently in the absence of the NA segment. In order to test for the
presence of such genetic determinants on the HA segment, we
attempted to rescue A/Moscow/10/99×A/Reunion/586/04
(M0×R2) 6:1 reassortant viruses carrying the HA segment from MO
in combination with the PB1, PB2, PA, NP, M, and NS segments from
R2 (5 R2:HA-MO), or the HA segment from R2 in combination with
the PB1, PB2, PA, NP, M, and NS segments from MO (5 MO:HA-R2).
The reverse genetics protocol was as described above. No hemagglu-
tinating activity and M-vRNA concentrations lower than 106copies/
ml were detected upon viral amplification on MDCK cells (Fig. 4, V1
and V2, gray bars). When the A/Moscow/10/99-derived NA plasmid
was added, viral titers for the 5-MO:R2-HA virus increased signifi-
cantly (Fig. 4, V1, hatched bar), whereas viral titers for the 5-R2:MO-
HA virus remained low (Fig. 4, V2, hatched bar). These data suggested
that the HA segment was necessary but not sufficient to restore the R2
To investigate further the genetic basis of the R2 phenotype, we
attempted to rescue additional 6:1 MO×R2 reassortant viruses,
carrying the R2-HA segment together with various combinations of
MO and R2 segments (V3 to V7 in Fig. 4). Addition of the M and NS
segments of R2 to the R2-HA segment was not sufficient to restore
the R2 phenotype, i.e. a significant viral production in the absence of
the NA segment (Fig. 4, V3, gray bar). In contrast, addition of the PB1,
PB2, PA and NS segments did restore the R2 phenotype (Fig. 4, V6,
gray bar), suggesting that at least one of the segments encoding
the polymerase subunits was necessary. Combination of the R2-PB1
segment with the R2-HA segment and the PB2, PA, NP, M and NS
segments from MO led to a significant viral production in the absence
of the NA segment, at levels similar to those observed for the R2
virus (Fig. 4, V7, gray bar). On the other hand, introduction of the MO-
PB1 segment in the R2 background prevented viral production in
the absence of the NA segment (Fig. 4, V4, gray bar), but not in the
presence of the NA segment (Fig. 4, V4, hatched bar). Our data
Fig. 4. R2/MO recombinant viruses titers. Virus titers were expressed as mean log10M copies/ml, standard errors (SE) from two or more experiments. “MO”: gene of A/Moscow/10/
99 H3N2, “R2”: gene of A/Reunion/586/04 H3NA−, “CPE + or −”: respectively detectable or nondetectable growth, “7+1”: gene composition identical to the seven segments
composition with the addition of the NA segment from A/Moscow/10/99 H3N2 (hatched bar).
V. Moules et al. / Virology 404 (2010) 215–224
indicated that a strong determinant of R2 phenotype was present of
the R2-PB1 segment. However, when tested alone in the MO back-
ground, the R2-PB1 segment did not restore the R2 phenotype (Fig. 4,
V5, gray bar). Taken altogether, these data established that the
combination of both the R2-HA and R2-PB1 segments was necessary
and sufficient to confer an MO×R2 reassortant virus the ability to
grow efficiently on MDCK cells in the absence of the NA segment.
Analysis of HA sequences of the H3NA− strains
Sequence alignement revealed that the A/Reunion/586/04 and A/
Lyon-CHU/26430/03 isolates differed from the antigenically related
H3N2 reference strains, A/Wellington/1/04 and A/Wyoming/3/03,
respectively, at five and seven positions, respectively (Fig. 5 and
Table 2). However, only the mutations P103Q and K173E were found
specific to A/Reunion/586/04 HA protein when compared to both
several 2004 clinical isolates and A/Moscow/10/99 H3N2 reference
strain. Moreover, only one mutation I260M was found specific to A/
Lyon-CHU/26430/03 when compared to both several 2003 clinical
isolates and A/Moscow/10/99 H3N2 reference strain.
Genetic analysis of PB1 determinants of the H3NA− strains
Four amino acids differ between the PB1 proteins from the A/
Reunion/586/04 and A/Lyon-CHU/26430/03 H3NA− viruses on one
hand, and the PB1 protein from the A/Moscow/10/99 H3N2 virus on
the other hand. These amino acid substitutions could potentially
contribute to the ability of H3NA− viruses to grow in the absence of
the NA segment. The K/E (H3NA−/H3N2) substitution at residue 188
is located within the nuclear localization signal of PB1 (Akkina et al.,
1987; Fodor and Smith, 2004). The substitutions, D/N, N/D and Q/R
(H3NA−/H3N2) substitutions at residues 617, 619 and 621, respec-
tively, are located within the interaction domain of PB1 with PB2
(Gonzalez et al., 1996).
In order to identify the residues determinant with respect to the
H3NA− phenotype, we characterized a new series of recombinant 6:1
reassortants which harbored the MO-PB1 segment, either wild-type
or mutated at one or several of residues 188, 617, 619 and 621, in the
R2 background. As mentioned earlier, the reassortant virus harboring
the wild-type MO-PB1 segment in the R2 background did not rep-
licate efficiently in the absence of the NA segment (Fig. 6, V4, gray
bar), but it did so in the presence of the NA segment (Fig. 6, V4,
hatched bar). Introduction of the single K188E substitution or the
three D617N, N619D and Q621R substitutions in MO-PB1 did not
allow the recovery of a replication competent A(H3NA−) virus (Fig. 6,
V4.1 and V4.2, gray bars). When the NA segment was added, the
K188E mutant virus was produced at high titers (about 1011copies of
M-vRNA/ml), whereas the triple 617–619–621 mutant was produced
at low titers (about 107.8logs copies of M-vRNA/ml), suggesting that
the triple mutation on its own was detrimental (Fig. 6, V4.1 and V4.2,
Fig. 5. Structural comparison of the RBS of different hemagglutinins. A. Comparison between HA A/Wellington/1/04 and HA A/Reunion/586/04. The HAs were modeled with the
program Swiss-Pdb Viewer (Guex and Peitsch, 1997) using the human H3 (PDB: 4hmg) structure as the template. The image rendering was done using VMD (Humphrey et al.,
1996). For clarity, only the RBS of the HA A/Wellington/1/04 is shown, and is represented as a gray cartoon diagram. The sialic acid appears in orange. The amino acid substitutions
are numbered and represented in licorice, highlighted in royal blue for A/Wellington/1/04 and red for A A/Reunion/586/04. B. Comparison between HA A/Wyoming/3/03 and HA
A/Lyon-CHU/26430/03. For clarity, only the RBS of the HA A/Wyoming/3/03 is shown, and is represented as a gray cartoon diagram. The sialic acid appears in orange. The amino
acid substitutions are numbered and represented in licorice, highlighted in royal blue for A/Wyoming/3/03 and red for A/Lyon-CHU/26430/03. All the amino-acid substitutions are
listed in Table 2.
HA amino acid sequence alignment. HA sequence alignement between A/Reunion/586/04, A/Lyon-CHU/26430/03, A/Wellington/1/04, A/Wyoming/3/03 and A/Moscow/10/99
StrainsResidue at amino acid position
103126 128145 159173186 189219226227 26032 150
V. Moules et al. / Virology 404 (2010) 215–224
hatched bars). Simultaneous introduction of the 188 and the triple
617–619–621 mutations on MO-PB1 led to an efficient viral pro-
duction in the absence of the NA segment, both when combined with
six R2-derived segments (Fig. 6, V4.3, gray bar) and when combined
with the R2-HA segment and five MO-derived segments (Fig. 6, V4.4,
gray bar). The effect of a unique substitution at residue 617, 619
or 621 of PB1 was tested, in the background of the MO-PB1:R2
reassortant virus and in combination with the K188E substitution.
All three mutant viruses grew efficiently when the NA segment was
added (Fig. 6, V4.5 to V4.7, hatched bards), but the only virus which
grew efficiently in the absence of the NA segment was the K188E+
N617D double mutant.
To investigate further the functional differences between the
PB1 proteins from A/Moscow/10/99 (MO) or A/Reunion/586/04
(R2), transient in vivo reconstitution of viral ribonucleoproteins was
performed in the presence of an H3 or an NS viral-like CAT RNA.
Transcription/replication of the viral-like reporter RNAs was deter-
mined by measuring CAT levels in extracts prepared from transfected
293T cells. As shown in Fig. 7, the levels of CAT measured with the R2
proteinswere 1 to 2 log higheras comparedto the MO proteins. When
R2-PB1 was expressed in association with the MO-PB2, -PA and -NP
proteins, or conversely when MO-PB1 was expressed in association
with the R2-PB2, PA and NP proteins, intermediate levels of CAT were
observed between the activities of MO- and R2-derived RNPs.
Fig. 6. R2/MO PB1 mutant recombinant viruses titers. Virus titers were expressed as mean log10M copies/ml, standard errors (SE) from two or more experiments. “MO”: gene of A/
Moscow/10/99 H3N2, “R2”: gene of A/Reunion/586/04 H3NA−, “CPE + or −”: respectively detectable or nondetectable growth, “7+1”: gene composition identical to the seven
segments composition with the addition of the NA segment from A/Moscow/10/99 H3N2 (hatched bar). Numbers in the PB1 box correspond to mutations realized in the PB1 gene.
Fig. 7. CATassays.graybars:NSviral-likeRNA(MO=100%);blackbars:H3viral-likeRNA.
V. Moules et al. / Virology 404 (2010) 215–224
Inthe present study, two influenza isolates, A/Lyon-CHU/26430/03
and A/Reunion/586/04, which showed no detectable sialidase activity
when tested along with 788 additional H3N2 isolates from the 2003
to 2005 period, as described previously (Ferraris et al., 2006), were
further examined. Upon characterization of plaque-purified progeny
viruses, we clearly established that the A/Lyon-CHU/26430/03 and A/
Reunion/586/04 lacked the integrality of the NA segment and prop-
agated a viral genome composed of seven segments. This property
is unique as compared to previously described NA-defective viruses,
which retain a deleted NA segment throughout serial amplifications on
cultured cells although they do not express a functional NA protein (Liu
and Air, 1993; Hughes et al., 2000). The ability of the A/Reunion/586/
04 virus, but not the H3N2 reference strain A/Moscow/10/99, to
replicate efficiently in the absence of any NA segment sequence was
confirmed using reverse genetics.
The growth kinetics of the A/Lyon-CHU/26430/03 and A/
Reunion/586/04 H3NA− viruses on MDCK and MDCK-SIAT1 cells
were delayed as compared to H3N2 reference strains. The end-point
viral yields were in the same range for H3NA− and H3N2 viruses on
MDCK cells whereas they were about 2–3-log lower for H3NA− as
compared to H3N2 viruses on MDCK-SIAT1 cells, in agreement with
the fact that higher amounts of Sialyl-α2,6-Galactose-containing
receptors at the surface of MDCK-SIAT1 cells increase viral depen-
dency on NA sialidase activity (Matrosovich et al., 2003). In addition
to the absence of sialidase activity, a reduced efficiency of genome
packaging probably contributes to the moderate growth defect of
H3NA− viruses. Indeed, upon cryo-EM analysis, an increased
proportion of particles with no visible ribonucleoproteins was
observed in H3NA− (about 25%) as compared to H3N2 viral stocks
(about 6%). The NA segmentsequences thatare usually retained in the
genome of NA-defective viruses correspond to the non-coding
sequences and sequences encoding the cytoplasmic tail and trans-
membrane region of the NA protein at the 3′ end, and the non-coding
sequences at the 5′ end (Yang et al., 1997). These sequences were
shown to be required for efficient incorporation of the viral genome in
budding particles (Fujii et al., 2003), which is in favor of the existence
of a selective mechanism underlying the concomitant packaging of a
set of eight genomic segments. The increased proportion of empty
particles in the H3NA− viral stocks most probably corresponds to a
packaging defect, due to the absence of NA packaging signals
(Hutchinson et al., 2010).
Published data suggest that the sequences present at the ex-
tremities of influenza genomic segments might also be required for
viral morphogenesis and stability (Yang et al., 1997; Hughes et al.,
2000; Enami and Enami, 1996). We observed no major morphological
defect of the H3NA− virions upon cryo-EM analysis as compared to
H3N2viruses. Both H3NA− andH3N2 virusesshowedvariableshapes
and sizes, in agreement with previously published observations (Booy
et al., 1985; Fujiyoshi et al., 1994; Harris et al., 2006). Most authors
agree that the NA/HA ratio at the surface of influenza A viruses is
about 1/5 (Harris et al., 2006; Fujiyoshi et al., 1994). The high reso-
lution of cryo-EM allowed us to measure the average spacing of
surface glycoproteins stalks and to extrapolate the surface glycopro-
teins density, precisely enough to detect a ∼20% reduction that could
result from the absence of NA tetramers . Interestingly, our data
revealed no marked difference in the surface glycoprotein density
between H3NA− and H3N2 viruses, which suggested that the surface
density of the H3 was up-regulated in H3NA− as compared to H3N2
viruses. This higher HA density could simply correspond to a steric
compensation of the absence of NA but the existence of other under-
lying mechanisms cannot be excluded.
Using reverse genetics, we established that both the HA and the
PB1 segments of the A/Reunion/586/04 H3NA− virus accounted for
its ability to replicate efficiently in the absence of the NA segment.
Mutations on the HA that reduce viral binding to sialic acids, and thus
reduce viral dependence on NA activity, have been found to confer
resistance to neuraminidase inhibitors (McKimm-Breschkin et al.,
1996). Such mutations on the HA of H3NA− viruses could also pos-
sibly provide a compensation for the absence of NA protein. HA
sequence alignement revealed that the A/Reunion/586/04 differed
from 2004 clinical samples at two positions. One of these, K173E
substitution, is located on protruding loops in the vicinity of the
receptor binding site, which are predicted to increase the negative
charges at the distal end of the HA, might alter the electrostatic
interactions between viral particles and the cell surface. One sub-
stitution, I260M, has been observed between A/Lyon-CHU/26430/
03 and 2003 clinical samples but no substitutions were specific to
H3NA− phenotype. In addition to these amino acid substitutions,
increased amounts of carbohydrates on the HA, due to the absence of
desialylation by the NA, may modulate the binding affinity of the HA
to the host receptors and thus may also contribute to the H3NA−
phenotype. Indeed, previous studies have shown that the removal of
sialic acids from oligosaccharides adjacent to the receptor binding
site of the HA by the NA protein increases the hemadsorption activity
of influenza viruses (Brassard and Lamb, 1997; Tong et al., 1998).
Here, we focused on the contribution of the PB1 segment of the A/
Reunion/586/04 isolate to the ability to grow in the absence of an NA
segment. The PB1 segment encodes three viral proteins. The PB1
protein is an RNA-dependent RNA polymerase and is the core subunit
of the heterotrimeric PB1-PB2-PA polymerase complex that ensures
transcription and replication of the viral genome in the nucleus of
infected cells (Elton et al., 2006). The PB1-F2 protein, which is en-
coded in an alternative reading frame in the A/Reunion/586/04 virus
but not in all influenza viruses, was found to be a pro-apoptotic and a
virulence factor (Chen et al., 2001), as well as a regulator of the viral
polymerase activity (Mazur et al., 2008). The functions of the third
most recently PB1-encoded protein, a truncated form of the PB1
protein (N40) are largely unknown (Wise et al., 2009). Sequence
alignement between PB1-F2 of H3NA− and H3N2 revealed no amino
and on reverse genetics experiments, we established that the nature
of both residues 188 and 617 of the PB1 protein (corresponding
to residues 148 and 577 of the N40 protein) were determinant with
respect to the ability of the A/Reunion/586/04 isolate to replicate
efficiently in the absence of the NA segment. The A/Reunion/586/04-
derived PB1 protein was also found to increase the efficiency with
which a viral-like RNA underwent transcription/replication upon
transient reconstitution of viral ribonucleoproteins, compared to the
A/Moscow/10/99-derived PB1 protein. The nature of residue 188,
which is located within a nuclear localization signal of PB1 (Akkina
et al., 1987; Fodor and Smith, 2004), could potentially have an impact
on the nuclear accumulation of PB1, whereas the nature of residue
617, which is located within the interaction domain of PB1 with
PB2, could modulate the assembly and/or stability of the polymerase
complex. Interestingly, a synergistic effect of substitutions at residues
188 and 617 was observed.
How does the observation of an increased polymerase activity
conferred by the PB1 protein of the R2 virus relate to its ability to
grow in the absence of the NA segment remains to be established.
One hypothesis is that an enhanced polymerase activity could
result in an increased expression of the HA protein, which would
allow viral attachment to the target cells despite a lower affinity
of the HA for sialic acids. Such an increased expression of the HA
could account for our observation that the density of glycopro-
teins at the surface of H3NA− virions was unchanged as
compared to H3N2 virions. This hypothesis is also in agreement
with recent reports of an enhanced polymerase activity being
correlated with an increased accumulation of the NA (Wanitchang
et al., 2010) or the HA protein (Marjuki et al., 2007) on the cell
surface. A second hypothesis is that an increased polymerase
V. Moules et al. / Virology 404 (2010) 215–224
activity could result in an increased production and accumulation
of viral ribonucleoproteins at the assembly site, which would
allow the incorporation of sets of seven genomic segments in
budding particles, despite an overall decrease in the packaging
efficiency due to the absence of the NA packaging signals (Fujii et
Although H3NA− viruses were isolated from nasal swabs in two
patients, the fact that they did not diffuse in the human population,
taken together with our observation of their growth defect on MDCK-
SIAT1 cells as compared to H3N2 viruses, suggest that these viruses
may replicate poorly in vivo. A bacterial co-infection, resulting in the
expression of a bacterial neuraminidase and favoring the multiplica-
tion of H3NA− variants, could not be established from the analysis of
nasal swabs, but cannot be excluded. Thus, it is still uncertain whether
or not the loss of the whole NA segment could represent a mechanism
of emergence of naturally NAI-resistant viruses able to diffuse in
the human population. Most interestingly, analysis of these H3NA−
viruses suggests that molecular features of the PB1 segment could
strongly modulate the HA/NA ratio at the cell surface and thus the
HA/NA balance. By doing so, the PB1 segment could have a significant
impact on the virulence, transmissibility and host-range of human
and animal influenza viruses. The finding, by us and by others, of an
impact of the PB1 segment on the amount of glycoproteins expressed
at the virions surface may lead to some improvement in the design
of influenza vaccines in the future (Wanitchang et al., 2010). Finally
our demonstration that H3NA− viruses can be grown in vitro effi-
ciently in the absence of exogenous NA could prove useful for the
development of influenza-based vectors expressing a heterologous
gene, and complementary to already existing approaches (Shinya
et al., 2004).
Materials and methods
Cells and viruses
Viruses were isolated on Madin-Darby canine kidney (MDCK) cells
from clinical throat or nasal swabs provided by the French national
influenza monitoring network GROG (Groupes Régionaux d'Observa-
tion de la Grippe, France) to the WHO Collaborating Center for
influenza (Southern France). They were subsequently amplified on
MDCK cells (two passages). The reference strains A/Moscow/10/99
H3N2 and A/California/07/04 H3N2 (kindly provided by Dr A. Hay,
NIMR/MRC) were amplified in the same conditions. MDCK cells were
in serum free Ultra-MDCK medium (Lonza) supplemented with 2 mM
L-glutamine (Sigma Aldrich), penicillin (225 units/ml) and strepto-
mycin (225 µg/ml) (Lonza). MDCK cells over-expressing the 2,6-
sialyltransferase (MDCK-SIAT1 cells) (Matrosovich et al., 2003) were
kindly provided by Professor HD Klenk (Institute of Virology, Philipps
University, Marburg, Germany). When indicated, Clostridium perfrin-
gens NA (Sigma) was added to the infection medium at a final con-
centration of 0.002 U/ml.
Cryo-electron microscopy (Cryo-EM), measurement of viral structure
and statistical analysis
Influenza viruses were purified by centrifugation on a 20% sucrose
cushion in phosphate-buffered saline. A droplet of viral preparation
was applied on holey carbon grids and vitrified in liquid ethane.
Electron micrographs were recorded under low-dose conditions at
liquid nitrogen temperature with a Technai Sphera LaB6 200KV mi-
croscope. Images were collected at 30,000× magnification with a
defocus range of 2 µm. The virion diameter and the glycoprotein (GP)
spacing were determined on Cryo-EM micrographs using the ImageJ
software (http://rsbweb.nih.gov/ij/). The GP spike surface density
was calculated as described previously (Yamaguchi et al., 2008;
Terrier et al., 2009). Briefly, the spacing distances between four
consecutive GP stalks were measured. Thus, the total GP number
along a 100 nm diameter virion surface was calculated considering
the total surface area (4piR2with R=50 nm) and the viral surface
corresponding to 4×4=16 GPs. Surface density data were subjected
to ANOVA statistical analysis (http://faculty.vassar.edu/lowry/
Ultrathin section electron microscopy
For ultrastructural analysis, infected cell cultures were fixed in
a 0.1 M sodium cacodylate, 2% glutaraldehyde solution at room
temperature. After three washes in 0.2 M sodium cacodylate buffer,
the samples were post-fixed in an osmium tetroxide solution, stained
with an aqueous solution of uranylacetate, dehydrated in a graded
series of ethanol at room temperature and embedded in epon. After
polymerization, ultrathin sections (65 nm thick) were cut with a
diamond knife and picked up onto copper grids (300 mesh). Post-
staining of the sections was performed on a Leïca Ultrostainer with a
lead citratesolution.After air drying, grids wereexamined on a Philips
CM 120 transmission electron microscope at an acceleration voltage
of 80 kV.
Viral growth kinetics, end point titration, and quantification of viral RNA
Eagle's minimum elementary medium (EMEM, Lonza) supplemented
with1 µg/mltrypsin(Rochediagnostics)andfurtherincubatedat34 °C.
Harvested supernatants were centrifuged at 1500×g for 10 min and
stored at −70 °C until analysis. End point-titration assays were
performed on confluent layers of MDCK cells in 96-well plates. Briefly,
50 µl of 10-fold serial dilutions of each virus was inoculated into four
replicate wells. The 96-well microplates were incubated at 34 °C and
the presence of cytopathic effects (CPE) was monitored 3 days later
under the microscope. The presence of virus in supernatants was also
confirmed by hemagglutination tests using guinea pig erythrocytes.
The TCID50per ml values were determined using the Reed and Muench
statistical method. The amounts of M viral genomic segment (M-
vRNA) were determined by real-time RT-qPCR as described previously
(Bouscambert Duchamp et al., 2010).
A/Moscow/10/99 and A/Reunion/586/04 recombinant viruses
were generated by reverse genetics as previously described (Hoffmann
et al., 2000). Briefly, viral RNA was extracted from infected-MDCK cell
culture supernatant using the QIAmp viral RNA minikit (Qiagen)
according to the manufacturer's instructions. Two-step RT-PCR was
using an influenza A universal RT primer (Uni-12 primer «3′-
AGCAAAAGCAGG-5′ », Eurogentec, Belgium) and segment-specific
primers described by Hoffmann et al. (2000). The resulting cDNAs
were cloned into the pHW2000 vector (Hoffmann et al., 2000). Sets of
eight recombinant pHW2000 plasmids were mixed with the Superfect
reagent (Qiagen) in Opti-MEM (GIBCO-BRL), according to the manu-
facturer's instructions, and added to293Tcells in six-well tissueculture
harvested and diluted 1/10 in EMEM medium supplemented with
TPCK-trypsin (1 μg/ml) to infect confluent layers of MDCK cells.
Fluorometric NA activity and inhibition tests
The neuraminidase activity was determined by a fluorometric
assay as described previously (Ferraris et al., 2005). Twenty five
microliters of the viral stocks was serially diluted twofold and
V. Moules et al. / Virology 404 (2010) 215–224
incubated with 50 µl of 200 µM 2-(4-methylumbelliferyl)-alpha-D-N- Download full-text
acetylneuraminic acid (4-MUNANA) substrate (Sigma) at 37 °C for
1 h. The enzymatic reaction was stopped by the addition of 150 µl of
50 mM glycine (pH 10.4). The fluorescence level of the released 4-
methylumbelliferone was determined at an excitation wavelength of
355 nm and an emission wavelength of 460 nm with a BMG Labtek
Transfections and CAT assays
The pPR-FluA-CAT and pPR-H3-CAT plasmids, which direct the
expression of influenza NS and H3 viral-like RNAs, respectively have
been described earlier (Labadie et al., 2007). Subconfluent mono-
layers of 293T cells in 12 well plates were transfected using the
FUGENE 6 reagent (Roche) according to the manufacturer instruc-
tions. Briefly, mixes of the pHW2000-PB1, -PB2, -PA, -NP (0.5, 0.5, 0.5,
1 μg) and the pPR-FluA-CAT or pPR-H3-CAT plasmid (0.5 μg) were
resuspended in 45 μl of OPTI-MEM medium (Invitrogen) with 5 μl of
FUGENE 6 and were distributed onto cells. Following 24 h of incu-
bation at 37 °C, cell extracts were prepared in 250 μl of the lysis buffer
provided with the CAT ELISA kit (Roche), and tested for CAT levels.
This procedure allows detection of 0.05 ng/ml CAT.
Marburg, Germany) for providing MDCK-SIAT1 cells. We also wish to
thank J Skehel for his helpful discussion.
G., 2002. Characterization of 2 influenza A(H3N2) clinical isolates with reduced
susceptibility to neuraminidase inhibitors due to mutations in the hemagglutinin
gene. J. Infect. Dis. 186, 1074–1080.
Akkina, R.K., Chambers, T.M., Londo, D.R., Nayak, D.P., 1987. Intracellular localization of
the viral polymerase proteins in cells infected with influenza virus and cells
expressing PB1 protein from cloned cDNA. J. Virol. 61, 2217–2224.
Booy, F.P., Ruigrok, R.W., van Bruggen, E.F., 1985. Electron microscopy of influenza
virus. A comparison of negatively stained and ice-embedded particles. J. Mol. Biol.
Bouscambert Duchamp, M., Casalegno, J., Gillet, Y., Frobert, E., Bernard, E., Escuret, V.,
Billaud, G., Valette, M., Javouhey, E., Lina, B., Floret, D., Morfin, F., 2010. Pandemic A
(H1N1)2009 influenza virus detection by real time RT-PCR: is viral quantification
interesting? Clin. Microbiol. Infect. 16, 317–321.
Brassard, D.L., Lamb, R.A., 1997. Expression of influenza B virus hemagglutinin containing
multibasic residue cleavage sites. Virology 236, 234–248.
J., Palese, P., Henklein, P., Bennink, J.R., Yewdell, J.W., 2001. A novel influenza A virus
mitochondrial protein that induces cell death. Nat. Med. 7, 1306–1312.
Dharan, N.J., Gubareva, L.V., Meyer, J.J., Okomo-Adhiambo, M., McClinton, R.C., Marshall,
S.A., St George, K., Epperson, S., Brammer, L., Klimov, A.I., Bresee, J.S., Fry, A.M.,
Oseltamivir-Resistance Working Group, 2009. Infections with oseltamivir-resistant
influenza A(H1N1) virus in the United States. JAMA 301, 1034–1041.
Elton, D., Digard, P., Tiley, L., Ortin, J., 2006. Structure and function of the influenza virus
RNP. Influenza Virology: Current Topics. Caister Academic Press, Wymondham, UK,
stimulate the membrane association of the matrix protein. J. Virol. 70, 6653–6657.
Ferraris, O., Lina, B., 2008. Mutations of neuraminidase implicated in neuraminidase
inhibitors resistance. J. Clin. Virol. 41, 13–19.
Ferraris, O., Kessler, N., Lina, B., 2005. Sensitivity of influenza viruses to zanamivir and
of neuraminidase inhibitors in clinical practice. Antiviral Res. 68, 43–48.
Ferraris, O., Kessler, N., Valette, M., Lina, B., 2006. Evolution of the susceptibility to
antiviral drugs of A/H3N2 influenza viruses isolated in France from 2002 to 2005.
Vaccine 24, 6656–6659.
Fodor, E., Smith, M., 2004. The PA subunit is required for efficient nuclear accumulation
of the PB1 subunit of the influenza A virus RNA polymerase complex. J. Virol. 78,
Fujii, Y., Goto, H., Watanabe, T., Yoshida, T., Kawaoka, Y., 2003. Selective incorporation of
influenza virus RNA segments into virions. Proc. Natl. Acad. Sci. USA 100, 2002–2007.
Fujiyoshi, Y., Kume, N.P., Sakata, K., Sato, S.B., 1994. Fine structure of influenza A virus
observed by electron cryo-microscopy. EMBO J. 13, 318–326.
Gonzalez, S., Zurcher, T., Ortin, J., 1996. Identification of two separate domains in the
influenza virus PB1 protein involved in theinteraction with the PB2 and PA subunits: a
model for the viral RNA polymerase structure. Nucleic Acids Res. 24, 4456–4463.
Gubareva, L.V., Matrosovich, M.N., Brenner, M.K., Bethell, R.C., Webster, R.G., 1998.
Evidence for zanamivir resistance in an immunocompromised child infected with
influenza B virus. J. Infect. Dis. 178, 1257–1262.
Gubareva, L.V., Webster, R.G., Hayden, F.G., 2001. Comparison of the activities of
zanamivir, oseltamivir, and RWJ-270201 against clinical isolates of influenza virus
and neuraminidase inhibitor-resistant variants. Antimicrob. Agents Chemother. 45,
Guex, N., Peitsch, M.C., 1997. SWISS-MODEL and the Swiss-PdbViewer: an environment
for comparative protein modeling. Electrophoresis 18, 2714–2723.
Harris, A., Cardone, G., Winkler, D.C., Heymann, J.B., Brecher, M., White, J.M., Steven, A.
C., 2006. Influenza virus pleiomorphy characterized by cryoelectron tomography.
Proc. Natl. Acad. Sci. USA 103, 19123–19127.
viruses lacking sialidase activity can undergo multiple cycles of replication in cell
culture, eggs, or mice. J. Virol. 74, 5206–5212.
Humphrey, W., Dalke, A., Schulten, K., 1996. VMD—visual molecular dynamics. J. Mol.
Graph. 14, 33–38.
Hutchinson, E.C., von Kirchbach, J.C., Gog, J.R., Digard, P., 2010. Genome packaging in
influenza A virus. J. Gen. Virol. 91, 313–328 Review.
Labadie, K., Dos Santos Afonso, E., Rameix-Welti, M.A., Van der Werf, S., Naffakh, N.,
2007. Host-range determinants on the PB2 protein of influenza A viruses control
the interaction between the viral polymerase and nucleoprotein in human cells.
Virology 362, 271–282.
Liu, C., Air, G.M., 1993. Selection and characterization of a neuraminidase-minus
mutant of influenza virus and its rescue by cloned neuraminidase genes. Virology
Marjuki, H., Yen, H.L., Franks, J., Webster, R.G., Pleschka, S., Hoffmann, E., 2007. Higher
polymerase activity of a human influenza virus enhances activation of the hemagglu-
tinin-induced Raf/MEK/ERK signal cascade. Virol. J. 4, 1–19.
Matrosovich, M., Matrosovich, T., Carr, J., Roberts, N.A., Klenk, H.D., 2003. Over-
expression of the alpha-2, 6-sialyltransferase in MDCK cells increases influenza
virus sensitivity to neuraminidase inhibitors. J. Virol. 77, 8418–8425.
influenza A virus protein PB1-F2 regulates viral polymerase activity by interaction
with the PB1 protein. Cell. Microbiol. 10, 1140–1152.
McKimm-Breschkin, J.L., Blick, T.J., Sahasrabudhe, A., Tiong, T., Marshall, D., Hart, G.J.,
Bethell, R.C., Penn, C.R., 1996. Generation and characterization of variants of NWS:
Neu5Ac2en. Antimicrob. Agents Chemother. 40, 40–46.
Mitnaul, L.J., Matrosovich, M.N., Castrucci, M.R., Tuzikov, A.B., Bovin, N.V., Kobasa, D.,
Kawaoka, Y., 2000. Balanced hemagglutinin and neuraminidase activities are
critical for efficient replication of influenza A virus. J. Virol. 74, 6015–6020.
Nedyalkova, M.S., Hayden, F.G., Webster, R.G., Gubareva, L.V., 2002. Accumulation
of defective neuraminidase (NA) genes by influenza A viruses in the presence
of NA inhibitors as a marker of reduced dependence on NA. J. Infect. Dis. 185,
of ribonucleoprotein complexes in influenza A virus particles. Nature 439, 490–492.
influenza virus mutants defective in neuraminidase. Virology 61, 397–410.
Roberts, N.A., 2001. Treatment of influenza with neuraminidase inhibitors: virological
implications. Philos. Trans. Biol. Sci. 353, 1895–1897.
Shinya, K., Fujii, Y., Ito, H., Ito, T., Kawaoka, Y., 2004. Characterization of a neuraminidase-
Terrier, O., Rolland, J.P., Rosa-Calatrava, M., Lina, B., Thomas, D., Moules, V., 2009.
Virus Res. 142, 200–203.
Tong,N.,Nobusawa, E., Morishita, M., Nakajima, S., Nakajima, K., 1998. M protein correlates
A (H1N1) virus. J. Gen. Virol. 79, 2425–2434.
Wagner, R., Matrosovich, M., Klenk, H.D., 2002. Functional balance between haemagglu-
tinin and neuraminidase in influenza virus infections. Rev. Med. Virol. 12, 159–166.
Wanitchang,A., Kramyu, J., Jongkaewwattana, A., 2010. Enhancement of reverse genetics-
derived swine-origin H1N1 influenza virus seed vaccine growth by inclusion of
indigenous polymerase PB1 protein. Virus Res. 147, 145–148.
Whitley, R.J., Hayden, F.G., Reisinger, K.S., Young, N., Dutkowski, R., Ipe, D., Mills, R.G.,
Ward, P., 2001. Oral oseltamivir treatment of influenza in children. Pediatr. Infect.
Dis. J. 127–133 Erratum in: Pediatr Infect Dis J. 2001. 20, 421.
Wise, H.M., Foeglein, A., Sun, J., Dalton, R.M., Patel, S., Howard, W., Anderson, E.C.,
Barclay, W.S., Digard, P., 2009. A complicated message: identification of a novel
PB1-related protein translated from influenza A virus segment 2 mRNA. J. Virol. 83,
Yamaguchi, M., Danev, R., Nishiyama, K., Sugawara, K., Nagayama, K., 2008. Zernike
phase contrast electron microscopy of ice-embedded influenza A virus. J. Struct.
Biol. 162, 271–276.
Yang, P., Bansal, A., Liu, C., Air, G.M., 1997. Hemagglutinin specificity and
neuraminidase coding capacity of neuraminidase-deficient influenza viruses.
Virology 229, 155–165.
V. Moules et al. / Virology 404 (2010) 215–224