A chicken influenza virus recognizes fucosylated α2,3 sialoglycanreceptors on the epithelial cells lining upper respiratory tractsof chickens

Takahiro Hiono a,c, Masatoshi Okamatsu a,c, Shoko Nishihara b,c, Sayaka Takase-Yoden b,c,Yoshihiro Sakoda a,c, Hiroshi Kida a,b,c,n

a Laboratory of Microbiology, Department of Disease Control, Graduate School of Veterinary Medicine, Hokkaido University, Sapporo,Hokkaido 060-0818, Japanb Department of Bioinformatics, Faculty of Engineering, Soka University, Hachioji, Tokyo 192-8577, Japanc Research Center for Zoonosis Control, Hokkaido University, Sapporo, Hokkaido 001-0020, Japan

a r t i c l e i n f o

Article history:Received 27 December 2013Returned to author for revisions25 January 2014Accepted 4 March 2014Available online 4 April 2014

Keywords:Avian influenza virusInterspecies transmissionHemagglutininReceptorFucosylation

a b s t r a c t

Influenza viruses recognize sialoglycans as receptors. Although viruses isolated form chickens prefer-entially bind to sialic acid α2,3 galactose (SAα2,3Gal) glycans as do those of ducks, chickens were notexperimentally infected with viruses isolated from ducks. A chicken influenza virus, A/chicken/Ibaraki/1/2005 (H5N2) (Ck/IBR) bound to fucose-branched SAα2,3Gal glycans, whereas the binding towards linearSAα2,3Gal glycans was weak. On the epithelial cells of the upper respiratory tracts of chickens, fucose-branched SAα2,3Gal glycans were detected, but not linear SAα2,3Gal glycans. The growth of Ck/IBR inMDCK-FUT cells, which were genetically prepared to express fucose-branched SAα2,3Gal glycans, wassignificantly higher than that in the parental MDCK cells. The present results indicate that fucose-branched SAα2,3Gal glycans existing on the epithelial cells lining the upper respiratory tracts of chickensare critical for recognition by Ck/IBR.

& 2014 Elsevier Inc. All rights reserved.

Introduction

Influenza viruses recognize sialoglycans on the host epithelialcells as their receptors. The binding specificities of the hemagglutinin(HA) of influenza viruses to sialoglycans vary depending on the hostspecies from which the virus was isolated: avian influenza virusespreferentially bind to glycans terminating with sialic acid α2,3galactose (SAα2,3Gal), whereas human influenza viruses preferglycans terminating with sialic acid α2,6 galactose (SAα2,6Gal) onthe cell surface (Rogers and D'Souza, 1989; Rogers and Paulson,1983). In addition, these specificities are consistent with receptordistributions in the host tissues: epithelial cells of the duck intestinepredominantly express SAα2,3Gal glycans (Ito et al., 1998, 2000),whereas epithelial cells of the human trachea predominantly expressSAα2,6Gal glycans (Shinya et al., 2006). In order to investigate thedistribution of SAα2,3Gal or SAα2,6Gal receptors in the host tissues,2 lectins, Maackia amurensis agglutinin (MAA) recognizing carbohy-drate structures of SAα2,3Galβ1,4GlcNAc or SAα2,3Galβ1,3GalNAc

(Kawaguchi et al., 1974; Konami et al., 1994), and Sambucus nigraagglutinin (SNA) recognizing SAα2,6Gal/GalNAc (Shibuya et al., 1987)have been used (Ito et al., 1998, 2000). However, previous reportshowed that these lectins recognize limited glycans: MAA do notrecognize carbohydrate structures of SAα2,3Galβ1,3GlcNAc orSAα2,3Galβ1,3Glc (Johansson et al., 1999). Therefore, they do notrecognize all SAα2,3/2,6Gal glycans.

Influenza A viruses of each of the known subtypes (H1 to H16and N1 to N9) have been isolated fromwater birds, especially frommigratory ducks (Fouchier et al., 2005; Kida et al., 1980; Kida andYanagawa, 1979). In addition, influenza virus genome-like RNA ofdistinct lineage were recently detected from bats (Tong et al., 2012,2013). The migratory ducks are, thus, the natural hosts forinfluenza viruses. Previous studies showed that chickens wererarely infected directly with viruses isolated from ducks, eventhough most influenza viruses isolated from chickens recognizeSAα2,3Gal glycans as the receptors, similarly to the viruses of duckorigin (Kishida et al., 2004; Matrosovich et al., 1997; Rogers andPaulson, 1983). Until now, inconsistent results regarding thedistribution of SAα2,3Gal and SAα2,6Gal in the epithelial cells ofthe chicken trachea have been reported (Costa et al., 2012;Kuchipudi et al., 2009; Pillai and Lee, 2010). Despite such areceptor specificity of chicken influenza viruses, it was reportedthat SAα2,6Gal glycans predominantly exist on the surface of

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/yviro

Virology

http://dx.doi.org/10.1016/j.virol.2014.03.0040042-6822/& 2014 Elsevier Inc. All rights reserved.

n Corresponding author at: Laboratory of Microbiology, Department of DiseaseControl, Graduate School of Veterinary Medicine, Hokkaido University, Kita 18Nishi 9, Kita-ku, Sapporo, Hokkaido 060-0818, Japan. Tel.: þ81 11 706 5207;fax: þ81 11 706 5273.

E-mail address: kida@vetmed.hokudai.ac.jp (H. Kida).

Virology 456-457 (2014) 131–138

epithelial cells lining the trachea of chickens (Kuchipudi et al.,2009). Gambaryan et al. (2008) reported that influenza virusesisolated from terrestrial birds, including chickens, preferentiallybind to 6-sulfo sialyl Lewis X [SAα2,3Galβ1,4(Fucα1,3)(6HSO3)GlcNAc], to which viruses from ducks do not bind. Thesemodifications of SAα2,3Gal glycans such as fucosylation or sulfa-tion appear to be important for the recognition of receptorsby influenza viruses. Recent developments in technology forglycoanalysis enable us to analyze the binding of influenza virusesincluding chicken influenza viruses to vast numbers of glycans(Belser et al., 2008; Stevens et al., 2006, 2008). However, theeffects of the structure of glycans other than the linkage ofterminal sialic acid and galactose on the receptor recognition ofinfluenza viruses have not been fully discussed. Detailed analysesof distribution of glycans on the human bronchial epithelial cellsand chicken erythrocytes revealed diversity of sialoglycan recep-tors on the surface of the host cells (Aich et al., 2011;Chandrasekaran et al., 2008). While, distribution of sialoglycanswhich are not recognized by MAA or SNA lectins in the chickenrespiratory tracts are not known. Consequently, the relationshipbetween glycan-binding specificity of influenza viruses isolatedfrom chickens and the distribution of virus receptors in chickens isnot well understood, although it is important for the control ofavian influenza in poultry. In the present study, to clarify the virusreceptor of chicken influenza viruses, we focused on the fucosyla-tion of SAα2,3Gal receptors. Furthermore, the distribution of thevirus receptor in the host tissues and binding specificity of thechicken influenza virus strain A/chicken/Ibaraki/1/2005 (H5N2)(Ck/IBR) to these receptors were analyzed.

Results

Experimental infection of chickens with an influenza virus of chickenor duck origin

In order to confirm whether Ck/IBR replicates in chickens, 105

PFU of Ck/IBR was intranasally inoculated into 4-week-old specificpathogen free (SPF) chickens (Table 1). An influenza virus of wildduck origin, A/duck/Mongolia/54/2001 (H5N2) (Dk/MNG), wasalso inoculated into chickens for control studies. Viruses wererecovered from the respiratory tracts of the chickens inoculatedwith Ck/IBR, whereas no virus was recovered from chickensinoculated with Dk/MNG. In order to investigate whether the HAprotein of influenza viruses, which plays roles of virus attachmentand entry to host cells, is responsible for the efficient replication ofCk/IBR in chickens, rgMNG/IBR-HA, having only HA gene derived

from Ck/IBR and the others from Dk/MNG, was generated andintranasally inoculated into chickens. As well as Ck/IBR, rgMNG/IBR-HA replicated in the upper respiratory tracts of the chickens.The results indicate that the HA is the major determinant forefficient growth of Ck/IBR in chickens.

Binding specificities of Ck/IBR and Dk/MNG to linearand fucose-branched SAα2,3Gal glycans

The results of experimental infection of chickens revealed thatvirus attachment to host receptors by the HA was responsible for

Table 1Virus recovery from the chickens intranasally inoculated with each virus.

Viruses Chicken number Swabs (log PFU/ml) Tissue samples (log PFU/g, 5 dpi)

Oral Cloacal Larynx Trachea Lung Kidney Colon

3 dpi 5 dpi 3 dpi 5 dpi

Ck/IBR 1 2.9 2.3 – – 4.0 – – – –

2 3.0 2.0 – – 3.0 – – – –

3 –a 1.7 – – – – – – –

Dk/MNG 4 – – – – – – – – –

5 – – – – – – – – –

6 – – – – – – – – –

rgMNG/IBR-HA 7 2.4 1.5 – – – – – – –

8 2.2 1.3 – – 4.1 2.7 – – –

9 2.0 2.0 – – 3.0 2.6 – – –

a Dash (–) indicates o1.0 for swabs and o2.0 for tissue samples.

0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Sialylglycopolymer (ng/well)

Abs

orba

nce

(450

-630

nm

)

0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0 50 100 150 200

0 50 100 150 200

Abs

orba

nce

(450

-630

nm

)

3 -SLN sLeX 6 -SL

Fig. 1. Receptor specificities of Ck/IBR and Dk/MNG. Binding of Ck/IBR (a) and Dk/MNG (b) to sialylglycopolymers containing 3'-SLN (black circles), sLeX (blacksquares), or 6'-SL (white circles) was investigated using solid-phase direct bindingassays. The data are presented as mean7S.E. of triplicate experiments. npo0.05 inthe comparison of 3'-SLN and sLeX.

T. Hiono et al. / Virology 456-457 (2014) 131–138132

the efficient growth of Ck/IBR in chickens. Therefore, bindingspecificities of Ck/IBR and Dk/MNG to glycan receptors wereinvestigated using a solid-phase direct binding assay. Both Ck/IBR and Dk/MNG specifically bound to SAα2,3Gal glycans, but notto 6'-SL, SAα2,6Gal glycan (Fig. 1a and b). Notably, Ck/IBR bound tosialyl Lewis X (sLeX), which is a fucose-branched SAα2,3Galglycans, whereas binding towards 3'-sialyllactosamine (3'-SLN),which is a linear SAα2,3Gal, was weak.

Fucosylated SAα2,3Gal glycans are expressed on the epithelial cellsof the upper respiratory tracts of chickens

MAA lectin and an anti-sLeX monoclonal antibody, KM93, weresubjected to the solid-phase direct binding assay to confirm theirbinding specificity to glycans prior to their use for immunofluores-cence staining. MAA lectin did not bind to sLeX, which is fucose-branched SAα2,3Gal glycan, but did to 3'-SLN, which is linearSAα2,3Gal glycan (Fig. 2a) In contrast, KM93 bound to sLeX, but didnot bind to 3'-SLN (Fig. 2b).

The distribution of linear and fucose-branched SAα2,3Galreceptors on the epithelial cells lining the upper respiratory tractsof chickens and colon of ducks were examined using MAA lectinand KM93. On the epithelial cells of the chicken larynx andtrachea, linear SAα2,3Gal glycans were not detected using MAAlectin, whereas fucose-branched SAα2,3Gal glycans were abun-dantly detected by KM93 (Fig. 3a). Linear SAα2,3Gal glycans werepredominantly detected on the epithelial cells of the duck colonusing MAA lectin as reported previously (Ito et al., 2000), whereasfucose-branched SAα2,3Gal glycans were not detected (Fig. 3b).

Fucosylation of SAα2,3Gal receptors recognized by KM93 iscatalyzed by α1,3/4 fucosyltransferase. So far, 6 fucosyltransferases,

fucosyltransferase 3, 4, 5, 6, 7, and 9 are known as α1,3/4fucosyltransferases in humans (Ikeda et al., 2001). Therefore, thecoding regions of each fucosylltransferase gene of human andchicken was phylogenetically analyzed (Fig. 4a). In chickens,3 glycosyltransferase genes, cFUT4, cFUT7, and cFUT9 genes werepredicted as orthologs of human fucosyltransferase genes, FUT4,FUT7, and FUT9, respectively. A fucosyltransferase gene of chickenwas identified as an ortholog of human FUT3, FUT5 and FUT6genes, and designated as cFUT3/5/6 gene. Among these glycosyl-transferases, human fucosyltransferase 3, 4, 5, 6, and 7 catalyzefucosylation of SAα2,3Gal glycans (Ikeda et al., 2001). In order toclarify whether these fucosyltransferase genes are expressed in theepithelial cells of the chicken trachea, mRNA expression of thecFUT3/5/6, cFUT4, and cFUT7 genes were investigated by reversetranscription polymerase chain reaction (RT-PCR, Fig. 4b). Of these3 fucosyltransferase genes, mRNA of cFUT3/5/6 gene was detected,indicating that SAα2,3Gal receptors existing on the epithelial cellslining the trachea of chickens can be fucosylated by chickenfucosyltransferase 3/5/6.

Accordingly, these results revealed that fucose-branchedSAα2,3Gal glycans are expressed on the epithelial cells of theupper respiratory tracts of chickens.

The effect of fucosylation of SAα2,3Gal receptor on the growthof the influenza virus of chicken origin

Glycan-binding specificity of Ck/IBR indicated that fucosylationof SAα2,3Gal receptors determines the attachment of Ck/IBR to thecell surface (Fig. 1a). In order to clarify whether fucosylation ofSAα2,3Gal receptors results in the up-regulation of growth of Ck/IBR, chicken cFUT3/5/6 gene was transfected into MDCK cells toestablish MDCK-FUT cells, which stably express fucose-branchedSAα2,3Gal glycans on their cell surfaces. We obtained 2 indepen-dent clones of MDCK-FUT cells, and they were designated asMDCK-FUT clone A32 or clone C33 cells respectively. Fig. 5 showsthe distribution of linear and fucose-branched SAα2,3Gal glycanson the cell surface of parental MDCK (a), MDCK-FUT clone A32 (b),clone C33 (c) cells, and MDCK cells which were transfected withempty pCXN2 vectors (MDCK-Empty cells; d). On the surface ofparental MDCK cells, linear SAα2,3Gal glycans were detected usingMAA, whereas not fucose-branched SAα2,3Gal glycans. On theother hand, overexpressed fucosyltransferase 3,5,6 catalyze fuco-sylation of SAα2,3Gal glycans, thereby, fucose-branched SAα2,3Galglycans were abundantly detected on the surface of MDCK-FUTcells of each clone. The receptor distribution on the cell surface ofMDCK-Empty cells was similar to that of parental MDCK cells.

Then, growth potential of Ck/IBR was compared in MDCK andMDCK-FUT cells (Fig. 6). MDCK, MDCK-FUT clone A32, clone C33,and MDCK-Empty cells were infected with Ck/IBR at the multiplicityof infection (M.O.I) of 0.1. Virus growth of Ck/IBR in MDCK-FUT cellsof each clone was significantly higher than that in parental MDCKcells or MDCK-Empty cells. At 24 h post infection, virus titers inMDCK-FUT cells of each clone were 5- to 13-fold higher than thoseof MDCK or MDCK-Empty cells. Accordingly, fucosylation of SAα2,3-Gal receptors enhanced virus growth of Ck/IBR.

Discussion

The relationship between receptor specificity of the viruses anddistribution of virus receptors in the host tissues are majordeterminants of host range and tissue tropism of the viruses.Understanding the mechanisms of interspecies transmission ofinfluenza viruses is essential for the control of avian and mamma-lian including human influenza. It is, thus, important to clarify themechanisms of recognition of receptors by influenza viruses.

0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0 50 100 150 200

Abs

orba

nce

(450

-630

nm

)

0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0 50 100 150 200

Abs

orba

nce

(450

-630

nm

)

**

*

*

**

*****

3 -SLN sLeX 6 -SL

Sialylglycopolymer (ng/well)

Fig. 2. Recognition of glycans by MAA lectin and the KM93 anti-sLeX antibody.Binding of MAA lectin (a) and the KM93 anti-sLeX antibody (b) to sialylglycopo-lymers containing 3'-SLN (black circles), sLeX (black squares), or 6'SL (white circles)was assessed using solid-phase direct binding assays. The data are presented asmean7SE of triplicate experiments. npo0.05 between 3'-SLN and sLeX.

T. Hiono et al. / Virology 456-457 (2014) 131–138 133

So far, distribution of SAα2,3Gal and SAα2,6Gal glycansdetected by MAA or SNA lectin in the tissues of several mammalsand birds have been reported (Ito et al., 1998, 2000; Kuchipudiet al., 2009; Shinya et al., 2006; Suzuki et al., 2000); however itwas also reported that not all α2,3 sialoglycans can be detectedusing MAA lectin (Konami et al., 1994). This is the first report thatfucose-branched SAα2,3Gal glycans, which are not recognized byMAA lectin, were detected in the epithelial cells of the upperrespiratory tracts of chickens (Fig. 3a). Some of the previousreports suggested that SAα2,6Gal glycans are predominantlypresent on the epithelial cells of the chicken trachea (Kuchipudiet al., 2009), though influenza viruses isolated from chickenspreferentially bind to SAα2,3Gal glycans (Matrosovich et al.,1997). Thus, our findings have important implications for theunderstanding of receptor usage of chicken Influenza viruses.

In the present study, we demonstrated that carbohydratestructures other than the linkage of terminal sialic acid andgalactose are also critical for the receptor recognition of influenzaviruses (Fig. 6). It was reported that sulfation or fucosylation ofSAα2,3Gal glycans affects the binding affinity of the HA to theseglycans (Gambaryan et al., 2005, 2008). However, effects of thesemodifications of SAα2,3Gal glycans on virus growth were notclarified. Fucosylation of receptors on the cell surface of MDCK-FUT cells enhanced binding affinity of Ck/IBR and virus receptors,which resulted in the enhancement of virus growth of Ck/IBR inMDCK-FUT cells (Fig. 6). Ck/IBR also replicated in parental MDCKcells, which do not express fucosylated SAα2,3Gal receptors on

their cell surface (Fig. 6). One explanation is that Ck/IBR utilizedreceptors which were not detected using MAA lectin or KM93. Onthe other hand, Ck/IBR grew better in MDCK-FUT cells than inMDCK cells, which means Ck/IBR prefer fucosylated receptorsas virus receptors. Meanwhile, the glycome of MDCK cells is notknown; therefore, receptors on the surface of MDCK cells, whichare recognized by Ck/IBR are still unclear. For the completeunderstanding of the receptor recognition and host range ofinfluenza viruses, further studies are needed.

Ck/IBR and rgMNG/IBR-HA, which possesses the HA derivedfrom Ck/IBR, efficiently replicated in chickens (Table 1). We alsogenerated rgIBR/MNG-HA, having only HA gene derived fromDk/MNG and the others from Ck/IBR. Virus was not recoveredfrom chickens intranasally inoculated with this virus, similar toDk/MNG, the donor of the HA (data not shown). These resultsindicate that the HA of Ck/IBR is a necessary and sufficient factorfor influenza viruses to replicate in chickens. On the other hand,the results of receptor binding of Dk/MNG does not explain thefact that Dk/MNG did not replicate in chickens, because both Ck/IBR and Dk/MNG strongly bound to sLeX glycans, which areexpressed on the epithelial cells of the upper respiratory tractsof chickens. The difference of receptor-binding affinity of Ck/IBRand Dk/MNG to sLeX glycans may result in the difference of virusgrowth of these viruses in chickens. However, some functions ofthe HA following the receptor binding such as virus entry ormembrane fusion may also be involved in the host specificity ofDk/MNG and Ck/IBR.

MAA

Anti-sLeX

Merged

Chicken Duck

ColonTracheaLarynx

Fig. 3. Detection of fucose-branched SAα2,3Gal glycans in the host tissues. Linear SAα2,3Gal glycans were detected using TRITC-labeled MAA (red), and fucose-branchedSAα2,3Gal glycans were detected using an anti-sLeX antibody (KM93: green) by immunofluorescence assay in the chicken larynx and trachea (a) or the duck colon (b). Nucleiwere stained with DAPI (blue). Original magnification was �200. Arrows indicate epithelial cells. Scale bars are 100 μm.

T. Hiono et al. / Virology 456-457 (2014) 131–138134

Previous studies of glycan microarray screenings of avianinfluenza viruses indicated that some of the influenza virusesisolated from chickens bound to fucose-branched SAα2,3Galglycans and some did not (Belser et al., 2008; Stevens et al.,2008). The present results indicate that fucosylation of SAα2,3Galreceptor is the key factor determining the receptor binding of Ck/IBR. However, many other factors in the topology of SAα2,3Galglycans may be involved in the binding of chicken influenzaviruses. In the present study, we found fucose-branched SAα2,3Galglycans on the epithelial cells lining the upper respiratory tracts ofchickens (Fig. 3). Due to the lack of information about distributionof SAα2,3Gal receptors which are not detected by MAA lectin, onthe surface of host cells, the effects of the structure of glycansother than the linkage of terminal sialic acid and galactose on thereceptor recognition of influenza viruses have not been fullyunderstood. Therefore, analyses of the receptor binding to glycanswhich exist on the surface of host cells is the base for under-standing the host range of influenza viruses. To understand themolecular mechanisms of interspecies transmission of influenzaviruses, it is necessary to clarify the carbohydrate structures of thereceptors which influenza viruses recognize. In the present study,

we revealed that a chicken influenza virus, Ck/IBR, recognizesfucosylated SAα2,3Gal glycans as virus receptors (Fig. 6). Thepresent results should further elucidate the understanding of thehost range and receptor specificity of influenza viruses.

Materials and methods

Viruses and cells

Low pathogenic avian influenza virus, Ck/IBR (Okamatsu et al.,2007), was kindly provided by National Institute of Animal Health,National Agriculture and Food Research Organization, Tsukuba,Ibaraki, Japan. Influenza virus, Dk/MNG, was isolated from fecalsamples of migratory ducks in Mongolia in 2001 (Sakoda et al.,2010). The viruses were propagated in 10-day-old embryonatedchicken eggs at 35 1C for 48 h, and the infectious allantoic fluidswere used as virus stocks. Madin-Darby canine kidney (MDCK)cells were maintained in Minimum Essential Medium (MEM,Nissui Pharmaceutical) supplemented with 0.3 mg/ml L-glutamine,100 U/ml penicillin G, 0.1 mg/ml streptomycin, 8 μg/ml gentamicinand 10% calf serum. Human embryonic kidney 293T cells weremaintained in Dulbecco's Modified Eagle's Medium (D-MEM:Life technologies) supplemented with 0.3 mg/ml L-glutamine,100 U/ml penicillin G, 0.1 mg/ml streptomycin, 8 μg/ml gentamicinand 10% fetal calf serum.

Reverse genetics

Eight genes from Ck/IBR and Dk/MNG each were cloned toproduce viruses by reverse genetics as described previously(Hoffmann et al., 2000, 2001; Soda et al., 2011). Ck/IBR, Dk/MNG,and rgMNG/IBR-HA (the HA gene from Ck/IBR and the other genesegments from Dk/MNG) were generated and used in the follow-ing experiments.

Plaque assay

Ten-fold dilutions of viruses were inoculated onto confluentmonolayers of MDCK cells and incubated at 35 1C for 1 h. Unboundviruses were removed, and the cells were washed with phosphate-buffered saline (PBS). The cells were then overlaid with MEMcontaining 0.7% Bacto-agar (Life technologies) and 5 μg/ml trypsinacetylated (Sigma Aldrich). After 48 h of incubation at 35 1C, cellswere stained with 0.005% neutral red. After 24 h, visible plaqueswere counted.

Experimental infection of chickens

Chickens (Gallus gallus, Boris Brown) were obtained fromHokkaido Chuo Shukeijou, Hokkaido, Japan. Four-week-old SPFchickens were infected with 105.0 PFU of viruses intranasally. On3 days post infection (d.p.i.), oral and cloacal swabs of the chickenswere collected. On 5 d.p.i., chickens were euthanized and oral andcloacal swabs, larynx, trachea, lungs, kidney and colon werecollected. To prepare a 10% suspension with MEM, tissue sampleswere homogenized using a Multi-Beads Shocker (Yasui Kikai).Infectivity titers of swabs and tissue samples were calculated byplaque assay. All infected chickens were kept in self-containedisolator units (Tokiwa Kagaku) at a BSL3 biosafety facility at theGraduate School of Veterinary Medicine, Hokkaido University,Japan. All animal experiments were performed according to theguidelines of the institutional animal care and use committee ofHokkaido University (approval number; 11-0087).

FUT3 (NM_000149.3)

FUT5 (NM_002034.2)

FUT6 (NM_000150.2)

cFUT3/5/6 (XM_004948742.1)

FUT4 (NM_002033.3)

cFUT4 (NM_001031487.1)

FUT7 (NM_004479.3)

cFUT7 (NM_001031476.1)

FUT9 (NM_006581.3)

cFUT9 (NM_001079502.1)100

71

100

100

100

69

79

0.1

500400300200100

Fig. 4. Expression of α1,3/4 fucosyltransferase genes in the epithelial cells of thechicken trachea. α1,3/4 fucosyltransferase genes of human and chicken were phylo-genetically analyzed (a). Complete cording regions of each fucosyltransferase genewere used for ML phylogenetic analysis. Horizontal distances are proportional to theminimum number of nucleotide differences required to join nodes and sequences.Numbers at each node indicate the confidence level in bootstrap analysis with 1000replications. Chicken α1,3/4 fucosyltransferase genes were detected by RT-PCR (b).Predicted PCR products are 266 (cFUT3/5/6), 361 (cFUT4), and 381 (cFUT7) base pairs,respectively. Specific PCR amplicons were detected in the cFUT3/5/6 gene.

T. Hiono et al. / Virology 456-457 (2014) 131–138 135

Solid-phase direct binding assay

Receptor binding specificity of the viruses was assessed usinga solid-phase direct binding assay with sialylglycopolymers

3'-Sialyllactosamine (3'-SLN)-PAA, Sialyl Lewis X (sLeX)-PAA, and6'-Sialyllactose (6'-SL)-PAA (Cosmo Bio Co., Ltd.) as describedpreviously (Shichinohe et al., 2013, Fig. 7). These were approxi-mately 30-kDa polymers containing 20% mol sugar linked to apolyacrylamide backbone. Each sialylglycopolymer was seriallydiluted and added to each well of a Universal-BIND™, 96 wellpolystyrene strip well microplate (Corning). Each well was blockedwith 1% bovine serum albumin (BSA) at room temperature for 1 h.After washing with PBS containing 0.05% Tween20 (PBST), asolution containing influenza viruses (16 HAU in PBS) was addedto each well and the plates were incubated at 4 1C for 12 h. Afterwashing, mouse anti-HA monoclonal antibodies [H3/2; previouslyestablished in our laboratory using A/duck/Pennsylvania/10218/1984 (H5N2)] were added to each well and the plates wereincubated at 4 1C for 2 h. The wells were then washed andincubated with goat anti-mouse IgG-HRP conjugate (Bio-Rad) at4 1C for 2 h. After washing, 100 μl of the substrates including0.5 mM 3,3'-tetramethylbenzidine (TMB) and 0.04% H2O2, wereadded to each well. After incubation at room temperature for10 min, the reactions were stopped using 50 μl of 2N H2SO4, andabsorbance at 450/630 nm was measured using a MULTISCAN JX(Thermo Fisher Scientific Inc).

Glycan-binding specificity of MAA lectin and the anti-sialylLewis X antibody (KM93, Kyowa Hakko KIRIN Co., Ltd.) wereassessed by solid-phase direct binding assay as described abovewith modifications. Briefly, each sialylglycopolymer was seriallydiluted and add to each well of Universal-BINDTM, 96 wellpolystyrene strip well microplate (Corning). Each well was blockedwith BSA for 1 h. After washing 5 times with PBST, a solutioncontaining FITC-labeled MAA (1:200 dilution; EY Laboratories;catalog number: F-7801-2) or KM93 antibody (1:500 dilution) wasadded to each well and the plates were incubated at roomtemperature for 1 h. After washing 10 times with PBST, sheepanti-FITC-HRP conjugate (SouthernBiotech; for FITC-labeled MAA)or goat anti-mouse IgM(μ)-HRP conjugate (American Qualex; forKM93 antibody) was added to each well and the plates wereincubated at room temperature for 1 h. After washing 10 times

MAA

Anti-sLeX

Merged

Parent FUTA32 FUTC33 Empty

Fig. 5. Receptor distribution on MDCK and MDCK-FUT cells. Linear SAα2,3Gal glycans were detected using TRITC-labeled MAA (red), and fucose-branched SAα2,3Gal glycanswere detected using an anti-sLeX antibody (KM93: green) by immunofluorescence assay on parental MDCK cells (a), MDCK-FUT cells clone A32 (b), clone C33 (c), and MDCK-Empty cells (d). Nuclei were stained with DAPI (blue). Original magnification was �200. Scale bars are 100 μm.

<1.5

2.0

3.0

4.0

5.0

6.0

7.0

8.0

0 4 8 12 16 20 24[h. p. i.]

Viru

s tit

er (L

og10

EID

50/m

l)

***

Parent

FUT A32 FUT C33

Empty

Fig. 6. Viral replication of Ck/IBR in MDCK or MDCK-FUT cells. MDCK cells (blackcircles), MDCK-FUT clone A32 cells (white circles), clone C33 cells (white squares), andMDCK-Empty cells (black squares) were infected with Ck/IBR at M.O.I. of 0.1. The dataare presented as mean7SE of triplicate experiments. npo0.05 between MDCK-FUTclone A32 and parental MDCK cells, nnpo0.05 between MDCK-FUT clone A32 andMDCK-Empty cells, †po0.05 between MDCK-FUT clone C33 and parental MDCK cells,and ††po0.05 between MDCK-FUT clone C33 and MDCK-Empty cells.

T. Hiono et al. / Virology 456-457 (2014) 131–138136

with PBST, 100 μl of premixed substrate including 0.5 mM TMBand 0.04% H2O2 was added each well. After incubation at roomtemperature for 5 min, the reactions were stopped with 50 μl of2N H2SO4, and absorbance at 450/630 nm was measured.

Detection of sialoglycans in chickens and ducks

Chickens (Boris Brown) were obtained from Hokkaido ChuoShukeijou, Hokkaido, Japan. Ducks (Anas platyrhynchos var. domes-ticus, Chelly Valley) were obtained from Takikawa Shinseien, Hok-kaido, Japan. The animals were euthanized, and their larynxes,tracheas and colons were collected. Tissue samples were embeddedin Tissue-Tek O.C.T. compound (Sakura Finetechnical Co., Ltd.), andfrozen in cold hexane. Sections of each tissue (8 μm) were cut with acryostat (CM1100; Leica Microsystems), air dried, and fixed for 5 minin cold acetone. Before staining, the sections were washed 3 timeswith PBS, and blocked with 1% BSA in PBS. Confluent monolayers ofMDCK or MDCK-FUT cells seeded on Chamber slide (SCS-008;Matsunami) were washed once with PBS, and fixed for 5 min incold acetone. Before staining, the sections were washed 3 times withPBS, and blocked with 1% BSA in PBS. To detect linear and fucose-branched SAα2,3Gal glycans, tissue sections and cells were incubatedwith 250 μl of the mixture of TRITC-labeled MAA (1:100 dilution; EYLaboratories; catalog number: R-7801-2) and anti-sLeX antibody(KM93, 1:250 dilution) at room temperature for 1 h. After beingwashed 3 times with PBS, the sections and cells were incubated with250 μl of FITC-labeled anti-mouse IgM antibody (1:200 dilution,Santa Cruz Biotechnology) at room temperature for 1 h. After beingwashed three times with PBS, sections and cells were counterstainedwith 4',6-diamino-2-phenylindole dihydrochloride (DAPI; DojindoMolecular Technologies). The tissue sections and cells were observedunder a fluorescence microscope (Axiovert 200; Carl Zeiss).

Phylogenetic analysis of α1,3/4 fucosyltransferase genes of humanand chicken

Sequence data of FUT3 (GenBank accession number:NM_000149.3), FUT4 (NM_002033.3), FUT5 (NM_002034.2), FUT6(NM_000150.2), FUT7 (NM_004479.3), FUT9 (NM_006581.3) of

human fucosyltransferase genes and cFUT3/5/6 (XM_004948742.1), cFUT4 (NM_001031487.1), cFUT7 (NM_001031476.1) and cFUT9(NM_001079502.1) of chicken fucosyltransferase genes wereobtained from public databases. These sequences were analyzedby the maximum likelihood (ML) method using MEGA 5.0 software(http://www.megasoftware.net). The reliability of phylogeneticinference at each branch node was estimated by the boot strapmethod with 1000 replications.

Detection of fucosyltransferase mRNA in the chicken trachea

Chickens were euthanized and their tracheas were collected. Thetracheas were washed 3 times with PBS and treated with Liberase DHResearch Grade (500 μg/ml; Roche) in PBS for 30 min at 37 1C tocollect epithelial cells. The suspension of the collected epithelial cellswas centrifuged at 3000g for 5 min and washed twice with PBS. TotalRNA of the cell pellet was extracted using a commercial kit (TRI-zolReagent, Life technologies) and reverse transcribed with Oligo(dT)20Primer (Life technologies) and M-MLV Reverse Transcriptase (Lifetechnologies). PCR amplification of a target sequence was performedwith primers 5'-CACCTCGAGAAAACTATGAG-3' and 5'-TTATACAAAC-CACTTAGACAG-3' for chicken cFUT3/5/6 gene, primers 5'-GGCCTTCGA-GAACTCCCAGC-3' and 5'-TCAGCTCTCAAACCAGCCGGC-3' for chickenFUT4 gene, and primers 5'-CCCACGACATCCAAGTCCAA-3' and 5'-CTAGGAGTTGAACCAGCTCTG-3' for chicken FUT7 gene. PCR wereperformed as follows: 95 1C for 5 min, then, 30 cycles of 94 1C for1 min, 50 1C for 1min, 72 1C for 40 s, and finally, 72 1C for 5 min. ThePCR amplicons were separated using electrophoresis in a 1.5% agarosegel, stained with ethidium bromide, and visualized under PrintgraphAE-6932 (ATTO).

Establishment of fucosyltransferase overexpressed MDCK cell

The 5' half of the chicken cFUT3/5/6 gene was amplified withKOD -Plus- Neo (Toyobo) using primers 5'-ATGCAGCCATCACT-CAGTGC-3' and 5'-CCATCTTGGTCTTGGGTGGG-3'. The 3' half ofthe chicken cFUT3/5/6 gene was amplified using primers 5'-CCAAGTCATTCTCCAAACTTA-3' and 5'-TTATACAAACCACTTAGA-CAG-3'. PCR amplicons were independently cloned into pGEM-TEasy Vector (Promega), and designated as cFUT3/5/6-5'half-pGEM-T and cFUT3/5/6-3'half-pGEM-T, respectively. cFUT3/5/6-5'half-pGEM-T and cFUT3/5/6-3'half-pGEM-T were digested with SacII(Toyobo) at 37 1C for 2 h, following the digestion with Bpu1102 I(Takara) at 37 1C for 11 h. The complete open reading frame of thechicken cFUT3/5/6 gene was constructed in pGEM-T Easy Vector byligation of the digested fragments using DNA Ligation KitoMighty Mix4 (Takara), and designated as cFUT3/5/6-pGEM-T.cFUT3/5/6-pGEM-T was digested with EcoRI (Toyobo) at 37 1C for2 h. Then, DNA fragments coding the complete chicken cFUT3/5/6gene were cloned into pCXN2 mammalian expression vector(Niwa et al., 1991; Suda et al., 2004) and designated cFUT3/5/6-pCXN2. MDCK cells were transfected with 1 μg of cFUT3/5/6-pCXN2 or empty pCXN2 vectors using Lipofectamine2000 (Lifetechnologies) and maintained in a culture medium supplementedwith 500 μg/ml of G418 sulfate (Calbiochem). To establish clonedMDCK-FUT cells, cFUT3/5/6-pCXN2 transfected cells were passaged7 times and then, G418-resistant clones were isolated and trans-ferred to 24-well plates. Each clone was passaged 3 additionallytimes in the presence of 500 μg/ml G418 sulfate and was frozen inaliquots. MDCK-FUT clone A32 and clone C33 cells were estab-lished from the cells batches independently transfected withcFUT3/5/6-pCXN2. To establish MDCK-Empty cells, cells trans-fected with empty pCXN2 vectors were passaged 7 times in thepresence of 500 μg/ml G418 sulfate and were frozen in aliquots.The gene transfected cells were maintained in the presence of500 μg/ml G418 sulfate, except when they were used for viral

PAA

PAA

PAA

SA Gal GlcNAcGlc Fuc

Fig. 7. Carbohydrate structures of sialylglycopolymers used in this study. Eachsymbol indicates sialic acid (SA, purple diamonds), galactose (Gal, yellow circles),glucose (Glc, blue circles), N-acetylglucosamine (GlcNAc, blue squares), and fucose(Fuc, red triangles), respectively. PAA indicates polyacrylamide.

T. Hiono et al. / Virology 456-457 (2014) 131–138 137

infection. The cells in these experiments were used at passagelevels from 3–10 starting from the frozen cell stock.

Virus growth in MDCK and MDCK-FUT cells

Confluent monolayers of MDCK, MDCK-FUT clone A32, cloneC33, and MDCK-Empty cells seeded on 6-well plate were infectedwith Ck/IBR at the M.O.I of 0.1. Four, 8, 12, and 24 h post infection,cell supernatants were harvested. Infectivity titers of the super-natants were calculated by the method of Read and Muench(1938), and expressed as EID50 per milliliter.

Acknowledgments

We thank M. Motoshima and N. Yamamoto for excellenttechnical support. The present work was supported in part bythe Japan Initiative for Global Research Network on InfectiousDiseases (J-GRID) and Japan Science and Technology Agency BasicResearch Programs. The present work was supported by Programfor Leading Graduate Schools (F01) from Japan Society for thePromotion of Science (JSPS).

References

Aich, U., Beckley, N., Shriver, Z., Raman, R., Viswanathan, K., Hobbie, S., Sasise-kharan, R., 2011. Glycomics-based analysis of chicken red blood cells providesinsight into the selectivity of the viral agglutination assay. FEBS J. 278,1699–1712.

Belser, J.A., Blixt, O., Chen, L.M., Pappas, C., Maines, T.R., Van Hoeven, N., Donis, R.,Busch, J., McBride, R., Paulson, J.C., Katz, J.M., Tumpey, T.M., 2008. Contempor-ary North American influenza H7 viruses possess human receptor specificity:Implications for virus transmissibility. Proc. Natl. Acad. Sci. USA 105,7558–7563.

Chandrasekaran, A., Srinivasan, A., Raman, R., Viswanathan, K., Raguram, S.,Tumpey, T.M., Sasisekharan, V., Sasisekharan, R., 2008. Glycan topology deter-mines human adaptation of avian H5N1 virus hemagglutinin. Nat. Biotechnol.26, 107–113.

Costa, T., Chaves, A.J., Valle, R., Darji, A., van Riel, D., Kuiken, T., Majó, N., Ramis, A.,2012. Distribution patterns of influenza virus receptors and viral attachmentpatterns in the respiratory and intestinal tracts of seven avian species. Vet. Res.43, 28.

Fouchier, R.A., Munster, V., Wallensten, A., Bestebroer, T.M., Herfst, S., Smith, D.,Rimmelzwaan, G.F., Olsen, B., Osterhaus, A.D., 2005. Characterization of a novelinfluenza A virus hemagglutinin subtype (H16) obtained from black-headedgulls. J. Virol. 79, 2814–2822.

Gambaryan, A., Yamnikova, S., Lvov, D., Tuzikov, A., Chinarev, A., Pazynina, G.,Webster, R., Matrosovich, M., Bovin, N., 2005. Receptor specificity of influenzaviruses from birds and mammals: new data on involvement of the innerfragments of the carbohydrate chain. Virology 334, 276–283.

Gambaryan, A.S., Tuzikov, A.B., Pazynina, G.V., Desheva, J.A., Bovin, N.V., Matroso-vich, M.N., Klimov, A.I., 2008. 6-sulfo sialyl Lewis X is the common receptordeterminant recognized by H5, H6, H7 and H9 influenza viruses of terrestrialpoultry. Virol. J. 5, 85.

Hoffmann, E., Neumann, G., Kawaoka, Y., Hobom, G., Webster, R.G., 2000. A DNAtransfection system for generation of influenza A virus from eight plasmids.Proc. Natl. Acad. Sci. USA 97, 6108–6113.

Hoffmann, E., Stech, J., Guan, Y., Webster, R.G., Perez, D.R., 2001. Universal primerset for the full-length amplification of all influenza A viruses. Arch. Virol. 146,2275–2289.

Ikeda, N., Eguchi, H., Nishihara, S., Narimatsu, H., Kannagi, R., Irimura, T., Ohta, M.,Matsuda, H., Taniguchi, N., Honke, K., 2001. A remodeling system of the 3'-sulfo-Lewis a and 3'-sulfo-Lewis x epitopes. J. Biol. Chem. 276, 38588–38594.

Ito, T., Couceiro, J.N., Kelm, S., Baum, L.G., Krauss, S., Castrucci, M.R., Donatelli, I.,Kida, H., Paulson, J.C., Webster, R.G., Kawaoka, Y., 1998. Molecular basis for thegeneration in pigs of influenza A viruses with pandemic potential. J. Virol. 72,7367–7373.

Ito, T., Suzuki, Y., Suzuki, T., Takada, A., Horimoto, T., Wells, K., Kida, H., Otsuki, K.,Kiso, M., Ishida, H., Kawaoka, Y., 2000. Recognition of N-glycolylneuraminic acidlinked to galactose by the alpha2,3 linkage is associated with intestinalreplication of influenza A virus in ducks. J. Virol. 74, 9300–9305.

Johansson, L., Johansson, P., Miller-Podraza, H., 1999. Detection by the lectins fromMaackia amurensis and Sambucus nigra of 3- and 6-linked sialic acid in

gangliosides with neolacto chains separated on thin-layer chromatogramsand blotted to PVDF membranes. Anal. Biochem. 267, 239–241.

Kawaguchi, T., Matsumoto, I., Osawa, T., 1974. Studies on hemagglutinins fromMaackia amurensis seeds. J. Biol. Chem. 249, 2786–2792.

Kida, H., Yanagawa, R., 1979. Isolation and characterization of influenza a virusesfromwild free-flying ducks in Hokkaido, Japan. Zentralbl. Bakteriol. Orig. A 244,135–143.

Kida, H., Yanagawa, R., Matsuoka, Y., 1980. Duck influenza lacking evidence ofdisease signs and immune response. Infect. Immun. 30, 547–553.

Kishida, N., Sakoda, Y., Eto, M., Sunaga, Y., Kida, H., 2004. Co-infection ofStaphylococcus aureus or Haemophilus paragallinarum exacerbates H9N2influenza A virus infection in chickens. Arch. Virol. 149, 2095–2104.

Konami, Y., Yamamoto, K., Osawa, T., Irimura, T., 1994. Strong affinity of Maackiaamurensis hemagglutinin (MAH) for sialic acid-containing Ser/Thr-linkedcarbohydrate chains of N-terminal octapeptides from human glycophorin A.FEBS Lett. 342, 334–338.

Kuchipudi, S.V., Nelli, R., White, G.A., Bain, M., Chang, K.C., Dunham, S., 2009.Differences in influenza virus receptors in chickens and ducks: implications forinterspecies transmission. J. Mol. Genet. Med. 3, 143–151.

Matrosovich, M.N., Gambaryan, A.S., Teneberg, S., Piskarev, V.E., Yamnikova, S.S.,Lvov, D.K., Robertson, J.S., Karlsson, K.A., 1997. Avian influenza A viruses differfrom human viruses by recognition of sialyloligosaccharides and gangliosidesand by a higher conservation of the HA receptor-binding site. Virology 233,224–234.

Niwa, H., Yamamura, K., Miyazaki, J., 1991. Efficient selection for high-expressiontransfectants with a novel eukaryotic vector. Gene 108, 193–199.

Okamatsu, M., Saito, T., Yamamoto, Y., Mase, M., Tsuduku, S., Nakamura, K.,Tsukamoto, K., Yamaguchi, S., 2007. Low pathogenicity H5N2 avian influenzaoutbreak in Japan during the 2005–2006. Vet. Microbiol. 124, 35–46.

Pillai, S.P., Lee, C.W., 2010. Species and age related differences in the type anddistribution of influenza virus receptors in different tissues of chickens, ducksand turkeys. Virol. J. 7, 5.

Read, L., Muench, H., 1938. A simple method of estimating fifty per cent endpoints.Am. J. Hyg. (Lond.) 27, 493–497.

Rogers, G.N., D'Souza, B.L., 1989. Receptor binding properties of human and animalH1 influenza virus isolates. Virology 173, 317–322.

Rogers, G.N., Paulson, J.C., 1983. Receptor determinants of human and animalinfluenza virus isolates: differences in receptor specificity of the H3 hemagglu-tinin based on species of origin. Virology 127, 361–373.

Sakoda, Y., Sugar, S., Batchluun, D., Erdene-Ochir, T.O., Okamatsu, M., Isoda, N., Soda,K., Takakuwa, H., Tsuda, Y., Yamamoto, N., Kishida, N., Matsuno, K., Nakayama,E., Kajihara, M., Yokoyama, A., Takada, A., Sodnomdarjaa, R., Kida, H., 2010.Characterization of H5N1 highly pathogenic avian influenza virus strainsisolated from migratory waterfowl in Mongolia on the way back from thesouthern Asia to their northern territory. Virology 406, 88–94.

Shibuya, N., Goldstein, I.J., Broekaert, W.F., Nsimba-Lubaki, M., Peeters, B., Peumans,W.J., 1987. The elderberry (Sambucus nigra L.) bark lectin recognizes theNeu5Ac(alpha 2-6)Gal/GalNAc sequence. J. Biol. Chem. 262, 1596–1601.

Shichinohe, S., Okamatsu, M., Sakoda, Y., Kida, H., 2013. Selection of H3 avianinfluenza viruses with SAα2,6Gal receptor specificity in pigs. Virology 444,404–408.

Shinya, K., Ebina, M., Yamada, S., Ono, M., Kasai, N., Kawaoka, Y., 2006. Avian flu:influenza virus receptors in the human airway. Nature 440, 435–436.

Soda, K., Asakura, S., Okamatsu, M., Sakoda, Y., Kida, H., 2011. H9N2 influenza virusacquires intravenous pathogenicity on the introduction of a pair of di-basicamino acid residues at the cleavage site of the hemagglutinin and consecutivepassages in chickens. Virol. J. 8, 64.

Stevens, J., Blixt, O., Chen, L.M., Donis, R.O., Paulson, J.C., Wilson, I.A., 2008. Recentavian H5N1 viruses exhibit increased propensity for acquiring human receptorspecificity. J. Mol. Biol. 381, 1382–1394.

Stevens, J., Blixt, O., Paulson, J.C., Wilson, I.A., 2006. Glycan microarray technologies:tools to survey host specificity of influenza viruses. Nat. Rev. Microbiol. 4,857–864.

Suda, T., Kamiyama, S., Suzuki, M., Kikuchi, N., Nakayama, K., Narimatsu, H., Jigami,Y., Aoki, T., Nishihara, S., 2004. Molecular cloning and characterization of ahuman multisubstrate specific nucleotide-sugar transporter homologous toDrosophila fringe connection. J. Biol. Chem. 279, 26469–26474.

Suzuki, Y., Ito, T., Suzuki, T., Holland, R.E., Chambers, T.M., Kiso, M., Ishida, H.,Kawaoka, Y., 2000. Sialic acid species as a determinant of the host range ofinfluenza A viruses. J. Virol. 74, 11825–11831.

Tong, S., Li, Y., Rivailler, P., Conrardy, C., Castillo, D.A., Chen, L.M., Recuenco, S.,Ellison, J.A., Davis, C.T., York, I.A., Turmelle, A.S., Moran, D., Rogers, S., Shi, M.,Tao, Y., Weil, M.R., Tang, K., Rowe, L.A., Sammons, S., Xu, X., Frace, M., Lindblade,K.A., Cox, N.J., Anderson, L.J., Rupprecht, C.E., Donis, R.O., 2012. A distinct lineageof influenza A virus from bats. Proc. Natl. Acad. Sci. USA 109, 4269–4274.

Tong, S., Zhu, X., Li, Y., Shi, M., Zhang, J., Bourgeois, M., Yang, H., Chen, X., Recuenco,S., Gomez, J., Chen, L.M., Johnson, A., Tao, Y., Dreyfus, C., Yu, W., McBride, R.,Carney, P.J., Gilbert, A.T., Chang, J., Guo, Z., Davis, C.T., Paulson, J.C., Stevens, J.,Rupprecht, C.E., Holmes, E.C., Wilson, I.A., Donis, R.O., 2013. New world batsharbor diverse influenza A viruses. PLoS Pathog. 9 (e1003657).

T. Hiono et al. / Virology 456-457 (2014) 131–138138