﻿FEBS Open Bio 2 (2012) 261­266
journal homepage: www.elsevier.com/locate/febsopenbio
Amino acid determinants conferring stable sialidase activity at low pH for H5N1
influenza A virus neuraminidase
Tadanobu Takahashia,b
, Chairul A. Nidomc
, Mai thi Quynh Led
, Takashi Suzukib,*
, Yoshihiro Kawaokaa,e,f,*
a
Influenza Research Institute, Department of Pathobiological Sciences, School of Veterinary Medicine, University of Wisconsin-Madison, Madison, WI, USA
b
Department of Biochemistry, School of Pharmaceutical Sciences, University of Shizuoka, and Global COE Program for Innovation in Human Health Sciences, Japan
c
Faculty of Veterinary Medicine, Tropical Disease Centre, Airlangga University, Surabaya, Indonesia
d
National Institute of Hygiene and Epidemiology, Hanoi, Viet Nam
e
Division of Virology, Department of Microbiology and Immunology and International Research Center for Infectious Diseases, Institute of Medical Science, University of Tokyo,
Shirokanedai, Minato-ku, Tokyo, Japan
f
ERATO Infection-Induced Host Responses Project, Japan Science and Technology Agency, Saitama, Japan
a r t i c l e i n f o
Article history:
Received 25 June 2012
Received in revised form 29 August 2012
Accepted 29 August 2012
Keywords:
Avian influenza A virus
H5N1
Highly pathogenic
Low-pH stability
Neuraminidase
Sialidase
a b s t r a c t
Avian influenza A viruses (IAVs) and human 1918, 1957, and 1968 pandemic IAVs all have neuraminidases
(NAs) that are stable at low pH sialidase activity, yet most human epidemic IAVs do not. We examined
the pH stability of H5N1 highly pathogenic avian IAV (HPAI) NAs and identified amino acids responsible
for conferring stability at low pH. We found that, unlike other avian viruses, most H5N1 IAVs isolated
since 2003 had NAs that were unstable at low pH, similar to human epidemic IAVs. These H5N1 viruses
are thus already human virus-like and, therefore, have the frequent infections of humans.
c 2012 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
1. Introduction
Since the first isolation of H5N1 highly pathogenic avian influenza
A virus (HPAI) from a 3-year-old boy in Hong Kong in 1997, more
than 600 confirmed cases of human H5N1 HPAI infection have been
reported, mainly from Indonesia, Viet Nam, and Egypt (as of June,
2012). Since 2004, frequent human infections have raised worldwide
concerns regarding the pandemic potential of H5N1 HPAI and its high
fatality rate (approximately 60% as of 2012 June) [1]. The viral enve-
lope glycoprotein hemagglutinin (HA) has been investigated exten-
sively and is essential to the high virulence and transmission of HPAI
to humans, because it possesses receptor binding ability (recogniz-
ing both avian- and human-type sialoglycoconjugates). It also has a
multi-basic cleavage site that facilitates systemic infection in chicken
Abbreviations: FBS, fetal bovine serum; HA, hemagglutinin; HPAI, highly pathogenic
avian influenza A virus; IAV, influenza A virus; NA, neuraminidase; PBS, phosphate-
buffered saline; TGF-, transforming growth factor-beta
* Takashi Suzuki; Department of Biochemistry, School of Pharmaceutical Sciences,
University of Shizuoka, Shizuoka 422-8526, Japan (T. Suzuki),Yoshihiro Kawaoka; In-
fluenza Research Institute, Department of Pathobiological Sciences, School of Veteri-
nary Medicine, University of Wisconsin, Madison, Wisconsin 53711, USA (Y. Kawaoka).
Tel.: +81 54 264 5725; fax: +81 54 264 5723 (T. Suzuki), Tel.: +1 608 265 4925; fax: +1
608 262 9641 (Y. Kawaoka).
E-mail address: suzukit@u-shizuoka-ken.ac.jp (T. Suzuki)
kawaokay@svm.vetmed.wisc.edu (Y. Kawaoka).
and mice [2,3]. Another viral envelope glycoprotein, neuraminidase
(NA), has sialidase activity that enhances virus release from the cell
surface by removing sialic acid from cellular glycoconjugates and viral
glycoproteins. Deletions in the NA stalk and a reduction in the level of
NA-induced transforming growth factor-beta (TGF-) influence the
high virulence of HPAI viruses [4,5]. However, the sialidase activity
of H5N1 HPAI NA per se has not been compared with that of human
influenza A virus (IAV) NA or other avian IAV NA [6].
We previously showed that influenza viruses differ in their sta-
bility at low pH. All avian IAV NAs tested to date are highly stable at
low pH; their sialidase activities are retained at pH 5.0 or less [7]. This
property may be associated with avian virus replication in intestines,
because for a virus to reach the intestine, it must pass through the
acidic environment of the gizzard. The NAs of pandemic human IAVs,
such as 1918 H1N1, 1957 H2N2, and 1968 H3N2 IAV, are also stable at
low pH. Viruses possessing a low-pH-stable NA from a pandemic IAV
in the background of A/WSN/33 (WSN; H1N1) replicated more effi-
ciently in cell culture and mouse lungs compared with a WSN virus
possessing a NA unstable at low pH [8]. This enhanced replicative
ability was associated with retention of the sialidase activity under
the acidic conditions of the endocytic pathway during virus entry [8].
On the other hand, the NAs of most seasonal human IAVs are unstable
at low pH [7,9,10].
Here, we examined the sialidase activity of 42 H5N1 HPAI viruses
2211-5463/$36.00 c 2012 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.fob.2012.08.007
262 Tadanobu Takahashi et al. / FEBS Open Bio 2 (2012) 261­266
isolated from humans, chickens, ducks, and other waterfowl and
found that most of the NAs, with the exception of three H5N1 viruses,
were not stable at low pH. We also identified amino acid determinants
that could confer low-pH stability to H5N1 HPAI NAs.
2. Materials and methods
2.1. Cells
Human embryonic kidney 293T cells were maintained in high
glucose Dulbecco's modified medium supplemented with 10% fetal
bovine serum (FBS).
2.2. NA genes and plasmids
All NA genes inserted into the pCAGGS expression vector and virus
abbreviations are listed in Supplementary Table 1. Chimeric NAs be-
tween A/duck/Guangdong/1/01 (H5N1) (DKG/1) and A/Hong Kong/
213/03 (H5N1) (HK/213) were generated by ligating the Mfe I and
BamH I sites (at positions 595 and 1145 in the NA gene, respectively)
or by ligating two PCR fragments derived from primers with the Bsa
I site (at position 842 in the NA gene). The Y155H, I289T, and Q313R
mutations of DKG/1 NA and the H155Y, T289I, and R313Q mutations
of HK/213 NA were introduced by means of PCR.
2.3. Sialidase activity of cell-expressed NA
293T cells (1.5 × 105 cells/well) in a 24-well tissue culture plate
were cultured overnight. The 70% confluent cells were transfected
with a plasmid (1 g/well) for NA expression by using TransIT-293
(Mirus, Madison, WI). After a 24-h incubation at 37 C, the trans-
fected cells were suspended in phosphate-buffered saline (PBS; 1.2
ml/well), and 50 l of each cell suspension was transferred into mi-
crotubes and centrifuged at 100 × g for 10 min. The cell pellets were
incubated with 57 l of 10 mM acetate buffer (pH 4.0, 5.0, or 6.0)
at 37 C for 10 min. Fifty microliters of each suspension was then
transferred to a 96-well black plate on ice and reacted with 2.5 l
of 2 mM 2 -(4-methylumbelliferyl)-N-acetylneuraminic acid (4MU-
Neu5Ac; Sigma-Aldrich Corp., St. Louis, MI) at 37 C for 30 min. The
reaction was stopped by the addition of 200 l of 100 mM sodium
carbonate buffer (pH 10.7). The fluorescent intensity (Ex, 355 nm;
Em, 460 nm) was measured with an Infinite M1000 microplate reader
(Tecan Group Ltd., M¨
annedorf, Switzerland). The sialidase activities of
the cell-expressed NAs were expressed as a percentage of the activity
at pH 6.0.
3. Results
3. 1. Low-pH stabilities of the sialidase activities of H5N1 HPAI NAs
We tested the low-pH stabilities of the sialidase activity of cells ex-
pressing each of the NA genes of 42 H5N1 HPAI viruses isolated from
human and avian hosts, as well as 10 other non-H5N1 avian IAVs
and eight human H1N1 IAVs. The NAs of all avian IAVs tested were
highly stable at low-pH, as described in our previous reports [7,9,10].
Three H5N1 HPAI NAs of A/Hong Kong/483/97 (HK/483), DKG/1, and
CKK/3, were also stable at low pH; at pH 4.0, their NAs had more than
30% of the sialidase activity at pH 6.0. Thirteen H5N1 HPAI NAs were
also moderately stable at low pH, retaining more than 30% of their
sialidase activity at pH 5.0 compared with that at pH 6.0. However,
the NAs of the remaining 26 H5N1 HPAIVs, which were isolated since
1997, were not low-pH-stable, like seasonal human IAVs. No distinct
host species-specific differences in their low-pH stabilities were ap-
parent among the H5N1 viruses isolated from different hosts such as
humans (Fig. 1A), aquatic birds (Fig. 1B), or chickens (Fig. 1C). There
were also no differences in the low-pH stabilities of H5N1 HPAI NAs
between aquatic and terrestrial birds (e.g., chickens). We also mea-
sured the sialidase activities of avian IAV NAs (Fig. 1D) and human
H1N1 IAV NAs (Fig. 1E) as controls of avian-like and human-like low-
pH stability. All avian IAV NAs tested were highly stable at low pH,
whereas most human H1N1 IAV NAs, except Spanish IAV A/Brevig
Mission/1/18 (H1N1) NA (BM/1), were not low-pH-stable, consistent
with our previous work [7,9,10].
When we measured the sialidase activity after pre-incubating the
NAs under acidic conditions, BM/1 NA lost its activity in a time-
dependent manner [10], whereas avian IAV NAs retained their siali-
dase activity even after 60 min of pre-incubation at pH 4.0 [10] (Sup-
plementary Fig. 1G and H). The low-pH-stable NAs of H5N1 HPAI
DKG/1 and A/chicken/Kyoto/3/04 (CKK/3) lost their sialidase activi-
ties, similarly to BM/1, as a function of increasing pre-incubation time
under acidic conditions (Supplementary Fig. 1A and C). The low-pH-
unstable H5N1 HK/213, A/whooper swan/Mongolia/6/05 (WSM/6),
and A/Viet Nam/1203/04 (VN/1203) NAs also lost their sialidase ac-
tivity, as did human H1N1 IAV NAs, in this assay (Supplementary
Fig. 1B, D, I, and J). These results indicate that most H5N1 HPAIs
have low-pH-unstable NAs, like seasonal human IAVs, and that H5N1
HPAIs with low-pH-stable NAs are the exception, like human pan-
demic Spanish IAVs, and their low-pH stability profiles differ from
the highly stable NAs at low pH in more common avian IAVs (Supple-
mentary Fig. 1).
3. 2. Identification of amino acid residues responsible for the low-pH
stability of an H5N1 HPAI NA
The amino acid sequence of the low-pH-stable DKG/1 NA was most
closely related to that of the low-pH-unstable HK/213 NA among the
viruses tested here (Supplementary Fig. 2). Therefore, to determine
the amino acid residues responsible for the low-pH stability of H5N1
HPAI NAs, we used the DKG/1 NA and the HK/213 NA to generate
chimeric NAs. There are eight amino acid differences between the
DKG/1 NA and the HK/213 NA in their ectodomains (amino acids 83­
468), at positions 155, 224, 257, 267, 289, 313, 341, and 415 (DKG/
1 NA numbering) (Table 1). First, we generated eight chimeric NAs
(Chimeras 1­8) by using the Mfe I and BamH sites and PCR primers
with the Bsa I site (Fig. 2A) and tested their low-pH stability. Based on
the properties of chimeras 3 and 4, residues at positions 155, 224, 257,
267, 289, 313 and 341 were identified as candidate determinants of
the low-pH stability of DKG/1 NA, whereas the results with chimeras
7 and 8 suggested a role for the amino acids at positions at 155, 289,
313, 341, and 415 in the low-pH stability of DKG/1 NA. Taken to-
gether, the residues at positions 155, 289, 313, and 341 appeared to
be determinants for the DKG/1 NA's low-pH stability (Fig. 2A, Table
1). We therefore generated an additional eight chimeric NAs in which
Tyr or His was substituted at position 155 in the different genetic
backbones (chimeras 9­16) (Fig. 2B). A striking difference in low-pH
stability between chimeras 13 and 14 suggested that the residues at
positions 289, 313, and 341 were likely determinants of low-pH sta-
bility. Finally, we generated three NAs with substitutions at positions
289, 313, and 341, respectively, in addition to the substitution at po-
sition 155 in the DKG/1 NA (Fig. 2C) or the HK/213 NA (Fig. 2D). The
dual substitutions at positions 155 and 341 substantially changed the
enzymatic stabilities at pH 4.0 and 5.0 (Table 1). We also examined
the low-pH stability of the HK/213 NA mutant with substitutions at
positions 155 and 341 after it was incubated under acidic conditions
in a time-dependent manner. The HK/213 NA with both the H155Y
and K341N mutations lost its sialidase activity in a time-dependent
manner, like the DKG/1 NA (Supplementary Fig. 1A and F), but it was
not as stable as the highly stable NAs at low pH in the more common
avian IAVs (Supplementary Fig. 1G and H). The DKG/1 NA with both
the Y155H and N341K mutations lost its sialidase activity within five
minutes of exposure to pH 4.0 and 5.0, like the HK/213 NA (Supple-
mentary Fig. 1B and E). Therefore, these results show that histidine at
Tadanobu Takahashi et al. / FEBS Open Bio 2 (2012) 261­266 263
Fig. 1. The low-pH stabilities of the sialidase activities of H5N1 HPAI NAs, human H1N1 IAV NAs, and non-H5N1 avian IAV NAs. Sialidase activities of NA-expressing cells transfected
with each NA gene were measured at pH 4.0 (filled columns), 5.0 (gray columns) and 6.0 (open columns). Sialidase activities are expressed as a percentage of each activity at pH
6.0. The NA genes used were derived from H5N1 HPAIs isolated from humans (A), aquatic birds (B), chickens (C), other avian IAVs (D) and from human H1N1 IAVs (E).
264 Tadanobu Takahashi et al. / FEBS Open Bio 2 (2012) 261­266
Fig. 2. Identification of amino acid residues responsible for the low-pH stability of DKG/1 NA and HK/213 NA. Sialidase activities were measured as described in the legend to Fig.
1. A, Low-pH stabilities of chimeric NAs. B, Low-pH stabilities of chimeric NAs with the Y155H or H155Y mutation at position 155. C, Low-pH stabilities of mutated DKG/1 NAs. D,
Low-pH stabilities of mutated HK/213 NAs.
Tadanobu Takahashi et al. / FEBS Open Bio 2 (2012) 261­266 265
Table 1
Amino acid comparison of the NA globular domains between DKG/1 and HK/213.
a
Amino acid position is based on DKG/1 NA numbering.b
Positions in the NA amino acid
sequence involved in creating the chimeric NAs.c
Boxed amino acids are responsible for
the low-pH stability of DKG/1 NA and HK/213 NA.
Fig. 3. Locations of the amino acid residues responsible for the low-pH stabilities of
H5N1 HPAI NAs. One subunit of the NA homotetramer structure (2HTY. pdb, VN/1203)
is shown in gray. In the surface model of NA, red and purple indicate the active site
and the calcium ion-binding site, respectively. The residues at positions 155 and 341
are colored in green and blue, respectively (DKG/1 NA numbering). Pictures were
generated by using the Pymol Molecular Graphics System Ver. 1.1r1 (DeLano Scientific
LLC).
position 155 and lysine at position 341 are responsible for the low-pH
instability of H5N1 HPAI NAs.
4. Discussion
Here, we showed that most H5N1 HPAI NAs are unstable at low
pH, like seasonal human IAV NAs. Thus, the low-pH stability of H5N1
HPAI NAs more closely resembles that of human IAV NAs than that
of avian IAV NAs, regardless of the host (i.e., humans and birds) from
which the virus is isolated. Although human infections with avian
H5N1, H7N3, H7N7, and H9N2 IAVs have occurred, by far the large
number of human infections has been reported for H5N1 HPAI [11].
Our previous research suggested that NA that are unstable at low pH
may be more suitable for the adaptation of IAV to humans [9]. The
low-pH instability of H5N1 HPAI NAs among avian IAV NAs might
contribute to the frequent transmission of H5N1 HPAI to humans.
For the DKG/1 NA and the HK/213 NA, both tyrosine at position
155 and asparagine at position 341 were responsible for the stability
at low pH; histidine at position 155 together with lysine at posi-
tion 341 destabilized NA activity at low pH. In the three-dimensional
structure of VN/1203 NA (Fig. 3) [12], the residue at position 155
was near the monomer interface, suggesting that it may be involved
in stabilization of the homotetrameric structure. The residue at po-
sition 341 was located near the calcium ion-binding site, which is
thought to be involved in the conformation of the enzymatic active
site [13,14]. We previously identified residues at positions 430, 435,
and 454 (BM/1 N1 numbering) as important for the low-pH stabil-
ity of N1 NAs (Supplementary Fig. 3A) [10] and residues at positions
344 and 466 (N2 numbering) as important for the low-pH stability of
N2 NAs (Supplementary Fig. 3B) [15]; these residues are also located
near the enzymatic active site, the calcium ion-binding site, and the
subunit interfaces.
Of the 42 H5N1 NAs tested, we found that the tyrosine at position
155 that is responsible for the low-pH stability of DKG/1 NA was
present in only three NAs [those of viruses HK/483, A/Hong Kong/
486/97 (HK/486), and DKG/1], whereas the remaining 39 H5N1 HPAI
NAs, including the low-pH-stable CKK/3 NA, had histidine at this
position, indicating that the low-pH stability of CKK/3 NA does not
require tyrosine at this position. The lysine at position 341 that was
responsible for the low-pH instability of HK/213 NA was present
only in this NA; the remaining 41 NAs tested here had asparagine
at this position. Therefore, these substitutions are not necessarily
responsible for the low-pH stabilities of all H5N1 HPAI NAs.
Several moderately stable NAs at low pH were found among the
42 H5N1 HPAI NAs (Supplementary Fig. 2). The amino acid sequences
of the low-pH-unstable A/Viet Nam/UT3028II/03 (VN/UT3028II) NA
and the moderately stable A/Viet Nam/UT3028/03 (VN/UT3028)
NA at low pH differ by only a single substitution at position 344
(DKG/1 NA numbering; cysteine for VN/UT3028II and glycine for
VN/UT3028). Also, only two amino acid differences were found be-
tween the unstable A/duck/Viet Nam/5001/04 (DKVN/5001) NA and
the moderately stable A/Viet Nam/UT30259/04 (VN/UT30259) NA at
low pH; the former having lysine and serine and the latter having
asparagine and glycine at positions 150 and 249 (DKG/1 NA number-
ing), respectively. These results indicate that amino acids at positions
other than 155 and 341 affect the low-pH stability of H5N1 HPAI NAs.
Among the residues responsible for the low-pH stability of the
NAs of H5N1 HPAI viruses, the residues at positions 249 and 344 are
located at or near the calcium ion-binding site, whereas the residues
at positions 150 and 249 are located near the enzymatic active site.
These results are consistent with our previous findings with BM/1 N1
and human N2 [10,15].
Unlike avian IAV NAs, most human IAV NAs are not stable at low
pH. Therefore, most H5N1 HPAI NAs isolated since 2003 are more
like the NAs of human IAVs in this regard, which might explain their
frequent infection of humans. Further studies of the low-pH stability
of IAV NAs may enable us to predict low-pH stability on the basis
of the amino acid residues located near the active site, the calcium
ion-binding site, and the subunit interfaces, which may, in turn, help
us to assess the adaptability of avian IAVs to humans.
Acknowledgments
We thank Susan Watson for scientific editing. This work was sup-
ported, in part, by MEXT/JSPS KAKENHI Grant Number (C; 23590549),
by a Sasakawa Scientific Research Grant from The Japan Science So-
ciety, by a SRI (Shizuoka Research Institute) academic research grant,
by the Global COE Program from the Japan Society for the Promotion
of Science, by ERATO (Japan Science and Technology Agency), by a
grant-in-aid for Specially Promoted Research from the Ministries of
Education, Culture, Sports, Science, and Technology, by grants-in-aid
from Health, Labor, and Welfare of Japan, by the Japan Initiative for
Global Research Network on Infectious Diseases, and by National In-
stitute of Allergy and Infectious Disease Public Health Service research
grants.
Supplementary Material
Supplementary material associated with this article can be found,
in the online version, at doi:10.1016/j.fob.2012.08.007.
References
266 Tadanobu Takahashi et al. / FEBS Open Bio 2 (2012) 261­266
[1] WHO (2012) Cumulative Number of Confirmed Human Cases of Avian
Influenza A/ (H5N1) Reported to WHO as of 7 June 2012. Avail-
able from: http://www.who.int/entity/influenza/human animal interface/
EN GIP 20120607CumulativeNumberH5N1cases.pdf.
[2] Basler C.F., Aguilar P.V. (2008) Progress in identifying virulence determinants of
the 1918 H1N1 and the Southeast Asian H5N1 influenza A viruses. Antiviral Res.
79, 166­178.
[3] Uchida Y., Chaichoune K., Wiriyarat W., Watanabe C., Hayashi T., Patchimasiri
T. et al. (2008) Molecular epidemiological analysis of highly pathogenic avian
influenza H5N1 subtype isolated from poultry and wild bird in Thailand. Virus
Res. 138, 70­80.
[4] Zhou H., Yu Z., Hu Y., Tu J., Zou W., Peng Y. et al. (2009) The special neu-
raminidase stalk-motif responsible for increased virulence and pathogenesis of
H5N1 influenza A virus. PLoS One. 4, e6277.
[5] Carlson C.M., Turpin E.A., Moser L.A., O'Brien K.B., Cline T.D., Jones J.C. et al.
(2010) Transforming growth factor-: activation by neuraminidase and role in
highly pathogenic H5N1 influenza pathogenesis. PLoS Pathog. 6, e1001136.
[6] Zhou J.J., Fu J., Fang D.Y., Yan H.J., Tian J., Zhou J.M. et al. (2007) Molecular
characterization of the surface glycoprotein genes of an H5N1 influenza virus
isolated from a human in Guangdong, China. Arch. Virol. 152, 1515­1521.
[7] Takahashi T., Suzuki Y., Nishinaka D., Kawase N., Kobayashi Y., Hidari K.I. et al.
(2001) Duck and human pandemic influenza A viruses retain sialidase activity
under low pH conditions. J. Biochem. 130, 279­283.
[8] Suzuki T., Takahashi T., Guo C.T., Hidari K.I., Miyamoto D., Goto H. et al. (2005)
Sialidase activity of influenza A virus in an endocytic pathway enhances viral
replication. J. Virol. 79, 11705­11715.
[9] Suzuki T., Takahashi T., Saito T., Guo C.T., Hidari K.I., Miyamoto D. et al. (2004)
Evolutional analysis of human influenza A virus N2 neuraminidase genes based
on the transition of the low-pH stability of sialidase activity. FEBS Lett. 557,
228­232.
[10] Takahashi T., Kurebayashi Y., Ikeya K., Mizuno T., Fukushima K., Kawamoto H.
et al. (2010) The low-pH stability discovered in neuraminidase of 1918 pandemic
influenza A virus enhances virus replication. PLoS One. 5, e15556.
[11] Wong S.S., Yuen K.Y. (2006) Avian influenza virus infections in humans. Chest.
129, 156­168.
[12] Russell R.J., Haire L.F., Stevens D.J., Collins P.J., Lin Y.P., Blackburn G.M. et al.
(2006) The structure of H5N1 avian influenza neuraminidase suggests new op-
portunities for drug design. Nature. 443, 45­49.
[13] Johansson B.E., Brett I.C. (2003) Variation in the divalent cation requirements of
influenza a virus N2 neuraminidases. J. Biochem. 134, 345­352.
[14] Brett I.C., Johansson B.E. (2006) Variation in the divalent cation requirements of
influenza A virus N1 neuraminidases. J. Biochem. 139, 439­447.
[15] Takahashi T., Suzuki T., Hidari K.I., Miyamoto D., Suzuki Y. (2003) A molecular
mechanism for the low-pH stability of sialidase activity of influenza A virus N2
neuraminidases. FEBS Lett. 543, 71­75.
