Virology 499 (2016) 170–177

Contents lists available at ScienceDirect

Virology

http://d0042-68

n CorrCollegeChina.

E-m

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

Identification and subcellular localization of porcine deltacoronavirusaccessory protein NS6

Puxian Fang a,b, Liurong Fang a,b, Xiaorong Liu a,b, Yingying Hong a,b, Yongle Wang a,b,Nan Dong a,b, Panpan Ma a,b, Jing Bi a,c, Dang Wang a,b, Shaobo Xiao a,b,n

a State Key Laboratory of Agricultural Microbiology, College of Veterinary Medicine, Huazhong Agricultural University, Wuhan 430070, Chinab The Cooperative Innovation Center for Sustainable Pig Production, Wuhan 430070, Chinac Department of Immunology and Aetology, College of Basic Medicine, Hubei University of Chinese Medicine, Wuhan 430065, China

a r t i c l e i n f o

Article history:Received 5 August 2016Returned to author for revisions14 September 2016Accepted 15 September 2016

Keywords:Porcine deltacoronavirusAccessory proteinNS6Subgenomic RNASubcelluar localization

x.doi.org/10.1016/j.virol.2016.09.01522/& 2016 Elsevier Inc. All rights reserved.

esponding author at: State Key Laboratoryof Veterinary Medicine, Huazhong Agricultura

ail address: vet@mail.hzau.edu.cn (S. Xiao).

a b s t r a c t

Porcine deltacoronavirus (PDCoV) is an emerging swine enteric coronavirus. Accessory proteins aregenus-specific for coronavirus, and two putative accessory proteins, NS6 and NS7, are predicted to beencoded by PDCoV; however, this remains to be confirmed experimentally. Here, we identified theleader-body junction sites of NS6 subgenomic RNA (sgRNA) and found that the actual transcriptionregulatory sequence (TRS) utilized by NS6 is non-canonical and is located upstream of the predicted TRS.Using the purified NS6 from an Escherichia coli expression system, we obtained two anti-NS6 monoclonalantibodies that could detect the predicted NS6 in cells infected with PDCoV or transfected with NS6-expressing plasmids. Further studies revealed that NS6 is always localized in the cytoplasm of PDCoV-infected cells, mainly co-localizing with the endoplasmic reticulum (ER) and ER-Golgi intermediatecompartments, as well as partially with the Golgi apparatus. Together, our results identify the NS6 sgRNAand demonstrate its expression in PDCoV-infected cells.

& 2016 Elsevier Inc. All rights reserved.

1. Introduction

Porcine deltacoronavirus (PDCoV) is an emerging swine entericcoronavirus that causes diarrhea in nursing piglets (Chen et al.,2015b; Hu et al., 2015; Jung et al., 2015; Ma et al., 2015). It was firstdetected in Hong Kong in 2012 (Woo et al., 2012). In early 2014,outbreaks of PDCoV were announced in swine populations in Ohio,Illinois, and Iowa, and it rapidly spread to multiple states in theUnited States (Li et al., 2014; Marthaler et al., 2014a; Marthaler et al.,2014b; Wang et al., 2014a, 2014b). Thereafter, PDCoVs were de-tected or caused outbreaks in Korea (Lee et al., 2016; Lee and Lee,2014), China (Dong et al., 2015; Song et al., 2015; Wang et al., 2015)and Thailand (Janetanakit et al., 2016; Madapong et al., 2016),posing significant economic concerns and gaining considerable at-tention (Jung et al., 2016; Lorsirigool et al., 2016; Zhang, 2016).

PDCoV is an enveloped, single-stranded, positive-sense RNAvirus belonging to the newly identified genus Deltacoronaviruswithin the family Coronaviridae. Its genome is approximately25.4 kb in length, making it the smallest genome among the knowncoronaviruses (CoVs). The genome arrangement of PDCoV is similar

of Agricultural Microbiology,l University, Wuhan 430070,

to those of other CoVs with the typical gene order: 5ʹUTR-ORF1a-ORF1b-S-E-M-NS6-N-NS7-3ʹUTR (Song et al., 2015; Woo et al.,2012). Although the biological functions of PDCoV-encoded proteinshave not been studied in detail, based on studies of other knownCoVs, PDCoV ORF1a and ORF1b probably encode two viral replicasepolyproteins, pp1a and pp1ab, which are predicted to be proteoly-tically cleaved into 15 mature nonstructural proteins responsible forviral replication and transcription; ORFs S, E, M, and N encode viralstructural proteins, Spike (S), Envelope (E), Membrane (M), andNucleocapsid (N), respectively (Chen et al., 2015a; Lee and Lee,2014; Li et al., 2014; Woo et al., 2010). Additionally, NS6 and NS7each encode a putative accessory protein (Fig. 1A).

Accessory proteins are genus-specific for coronavirus (Tan et al.,2006); however, different CoVs contain different numbers of ac-cessory genes and proteins. For example, alphacoronavirus felineinfectious peritonitis virus encodes five accessory proteins (p3a,p3b, p3c, p7a, and p7b), while only one accessory protein is en-coded by porcine epidemic diarrhea virus (PEDV); betacoronavirussevere acute respiratory syndrome coronavirus (SARS-CoV) en-codes a total of eight accessory proteins; and the most studiedgammacoronavirus, infectious bronchitis virus (IBV), encodes fouraccessory proteins (Liu et al., 2014). Although coronavirus acces-sory proteins have generally been considered to be dispensable forviral replication in vitro (Haijema et al., 2004; Yount et al., 2005),extensive functional studies have shown that many accessory

Fig. 1. Analysis of the leader-body TRS junction of putative PDCoV NS6 sgRNA. (A) The primer design for the leader-body junction RT-PCR analysis is shown in a schematicdiagram of the PDCoV full-length genome. (B) A representative gel from agarose gel electrophoresis of RT-PCR products amplified from PDCoV mRNA is shown. M, molecularsize ladder. (C) Analysis of the sgRNA NS6 sequence. The primers and the leader sequences are displayed as underlined and as italicized, respectively. The positions of thenucleotides in the genome sequences are indicated by black arrowheads. The start codon ATG in NS6 sgRNA is marked in bold. Boxed regions represent the authentic TRS andputative TRS used for sgRNA synthesis. The N100 indicates that the 100 nucleotides at that region are not shown.

P. Fang et al. / Virology 499 (2016) 170–177 171

proteins are involved in immune modulation (Kopecky-Bromberget al., 2007) and viral pathogenesis in vivo (De Haan et al., 2002).The field of coronavirus accessory proteins has gained consider-able attention in recent years.

In the PDCoV genome, there are two putative accessory genes,NS6 and NS7. NS6 is predicted to be located in the genome be-tween M and N and to encode a 94-amino acid peptide, while NS7is predicted to be located within the N gene (Lee and Lee, 2015;Woo et al., 2012). To date, there are no reports regarding the

expression of PDCoV accessory genes or the identification of anassociated transcription regulatory sequence (TRS) for productionof these subgenomic RNAs (sgRNAs) in virus-infected cells.

Here, we identified the leader-body fusion site and TRS of NS6sgRNA. By using monoclonal antibodies (MAbs) that recognize theputative NS6 protein, we demonstrated that the predicted NS6could be expressed and localized to the cytoplasm in PDCoV-in-fected cells, providing the first biochemical evidence for the ex-istence of PDCoV NS6.

Fig. 2. Expression and purification of the NS6 protein. (A) A representative imagefrom SDS-PAGE analysis of NS6 expression in E. coli after induction by IPTG. M,Protein molecular weight marker; lanes 1–4, the expressions of pET28a-NS6 afterinduction with IPTG for 3, 4, 5, and 6 h, respectively; lane 5, the expression ofpET28a-NS6 without IPTG induction; lane 6, the expression of empty vector pET-28a after induction with IPTG. (B) A representative image from a western blottinganalysis of expression products of pET28a-NS6 (lane 1) and empty vector pET-28a(lane 2) using anti-His MAb is shown. (C) A representative image from SDS-PAGEanalysis of purified recombinant NS6 protein is shown. Rosetta (DE3) cells har-boring pET28a-NS6 were induced by IPTG, and cells were collected at 3 h afterinduction and subjected to supersonic schizolysis. The supernatants (lane 1) andprecipitates (lane 2) of bacterium solution were collected. The precipitates werepurified (lane 3). M, Protein molecular weight marker.

Table 1Sequence of the primers used for identification of sgRNA and construction ofplasmids.

Primer Nucleotide sequencea (5′–3′) Nucleotide positionb

Leader-F AATTTTATCTCCCTAGCTTCG 40–60NS6r AAGAATGACTTGGAGCGATGGTCGA 23936–23960NS6-F ACTGAATTCATGTGCAACTGCCATCTGCAG 23694–23714NS6-R GCGCTCGAGATTTAATTCATCTTCAAGAATG 23954–23975

a Restriction enzyme recognition sites are shown in bolded and underlinedtype.

b The number represents nucleotide position corresponding to the nucleotidesequence of the PDCoV CHN-HN-2014 (GenBank accession no KT336560).

P. Fang et al. / Virology 499 (2016) 170–177172

2. Results

2.1. Identification of PDCoV NS6 sgRNA

A vital feature of coronavirus transcription is the set of sgRNAsproduced by discontinuous transcription. Each sgRNA contains acommon 5ʹ leader sequence derived from the 5ʹ end of the viralgenome and a so-called “body” sequence, which represents se-quences from the 3ʹ-poly(A) stretch to a position that is upstream ofeach genomic ORF encoding a structural or accessory protein (Sa-wicki et al., 2001, 2007). The fusion of leader and body sequences isprocessed, at least in part, by cis-acting elements termed tran-scription regulatory sequences (TRSs) (Hussain et al., 2005; Sawickiet al., 2007). Thus, determining the junction region of the leaderand body sequences can verify the existence of the correspondingsgRNAs. To identify the possible NS6 sgRNA, total intracellular RNAwas extracted from LLC-PK cells infected with PDCoV, and sgRNAswere amplified by leader-body junction RT-PCR with the primersLeader-F and NS6r (Fig. 1A) as reported previously (Dye and Siddell,2005; Hussain et al., 2005). As shown in Fig. 1B, two specific RT-PCRproducts of approximately 1.0 kb and 500 bp were obtained andthen isolated from the agarose gel, followed by cloning and se-quencing. At least 10 independent clones were sequenced. The re-sults reveal that the two specific RT-PCR products are sgRNAs M andNS6 (Fig. 1B). Our sequence analysis of these RT-PCR products in-dicates that sgRNA M contains a leader sequence followed by thetypical PDCoV TRS (ACACCA) (Woo et al., 2009, 2012), as expectedfor the predicted M sgRNA transcript. However, to our surprise, theleader-body fusion site (ACACCT) for sgRNA NS6 is 148 nucleotides,rather than the predicted 46 nucleotides, upstream of the AUG startcodon of the NS6 gene (Fig. 1C). Furthermore, there is a nucleotidedifference (underlined) in the TRS sequence (ACACCT) of NS6compared with that of the M gene (ACACCA), indicating that theTRS utilized by the NS6 gene is non-canonical.

2.2. Prokaryotic expression and purification of NS6 and generation ofanti-NS6 monoclonal antibodies (MAbs)

To further investigate the existence of NS6 at the protein level,we needed to prepare an NS6-specific antibody. To this end, a 285-bp cDNA fragment of the NS6 gene was amplified by RT-PCR andcloned into a prokaryotic expression vector, pET28a(þ), resulting inpET28a-NS6. This plasmid was then transformed into Escherichiacoli Rosette (DE3), and gene expression was induced with 0.8 Mmisopropylthiogalac-topyranoside (IPTG). A sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis showedthat the fusion protein His-NS6 was efficiently expressed in theform of inclusion bodies in E. coli, the size of which was consistentwith the predicted molecular weight of the recombinant protein(Fig. 2A). The results of western blotting assays demonstrated thatthe recombinant NS6 protein was specifically recognized by anti-HisMAb (Fig. 2B). Subsequently, the protein was purified (Fig. 2C), andthe concentration of the purified NS6 protein was 0.46 mg/mL, asmeasured by using a Trace ultraviolet spectrophotometer.

Using the purified NS6 protein as an immunization antigen and thehybridoma technique, two hybridoma cell lines, named 4B9 and 2G3,were acquired through screening and subcloning. To further confirmthe specificity of these two MAbs, a eukaryotic expression plasmidencoding NS6 was constructed by cloning the cDNA of the NS6 geneinto pCAGGS with the primers NS6-F and NS6-R (Table 1), which wasthen transfected into LLC-PK cells; indirect immunofluorescence as-says (IFAs) and western blotting assays were then performed with theobtained MAbs. As shown in Fig. 3A, specific fluorescence could beobserved in cells transfected with pCAGGS-NS6, while neither MAbrecognized cells transfected with the empty vector, pCAGGS. The re-sults of western blotting assays also show that approximately 11-kDa

specific protein bands, the size of which is identical to the size of thepredicted NS6 protein (Fig. 3B), could be detected by both MAbs in thelysates of cells transfected with pCAGGS-NS6.

Fig. 3. Expression of NS6 in cells transfected with pCAGGS-NS6. LLC-PK cells were platedonto 24-well plates following transfection with pCAGGS-NS6 or empty vector. At 28 hpost-transfection, cells were fixed for use in an IFA (A) with MAbs (4B9 or 2G3) againstNS6 and Alexa Fluor 594-conjugated donkey anti-mouse IgG (red). DAPI (blue) indicatesthe locations of the cell nuclei. Fluorescent images were acquired with a confocal laserscanning microscope (Fluoview ver. 3.1; Olympus, Japan). The transfected cells were alsocollected for use in western blotting (B) analyses performed with MAbs 4B9 or 2G3.

Fig. 4. Expression of NS6 in PDCoV-infected LLC-PK cells. LLC-PK cells were platedonto 24-well plates and then mock-infected or infected with PDCoV at a MOI of 5.0.At 12 h post-infection, IFAs (A) and western blotting assays (B) were performedwith MAbs 4B9 or 2G3, as described in Fig. 3.

P. Fang et al. / Virology 499 (2016) 170–177 173

2.3. Expression and subcellular localization of NS6 protein in PDCoV-infected cells

To test if NS6 is indeed expressed during PDCoV infection, LLC-PKcells were infected with PDCoV at a multiplicity of infection (MOI) of5.0. The cells were fixed and either subjected to IFA or collected forwestern blotting at 12 h post-infection (hpi). For IFA, obvious im-munofluorescence was observed in the cytoplasm in PDCoV-infectedLLC-PK cells, while no fluorescence signals were observed in mock-infected cells (Fig. 4A). Also, specific protein bands similar to the size

observed in pCAGGS-NS6-tranfected cells could be detected by bothMAbs in lysates of PDCoV-infected LLC-PK cells (Fig. 4B).

To further investigate the expression kinetics of NS6 protein indetail during PDCoV infection, LLC-PK cells seeded onto microscopecoverslips in 24-well plates were mock-infected or infected withPDCoV at a MOI of 10. At 3, 6, 9, 12, and 15 hpi, cells were fixed for IFAwith MAb 4B9. As shown in Fig. 5A, specific fluorescence could beobserved at 6 hpi, and increasingly more fluorescence was readilydetected as the infection progressed. Up to 15 hpi, obvious cytopathiceffects characterized by cell shrinkage and detachment were observed.

Fig. 5. Expression kinetics and subcellular localization of NS6 protein in PDCoV-infected LLC-PK cells. LLC-PK cells were plated onto coverslips and then mock-infected or infected withPDCoV at a MOI of 10. (A) At the indicated times post-infection, cells were fixed for IFAwithMAb 4B9 and Alexa Fluor 488-conjugated donkey anti-mouse IgG (green). (B) At 12 hpi,cells were fixed and stained with NS6-specific MAb (4B9) and rabbit anti-GRP94 (ER marker), rabbit anti-GS28 (Golgi marker), or rabbit anti-SEC31(ER-Golgi intermediatecompartment), followed by staining with secondary antibodies Alexa Fluor 488-conjugated donkey anti-rabbit IgG (green) or Alexa Fluor 594-conjugated donkey anti-mouse IgG(red). Nuclei were counterstained with DAPI (blue). Fluorescent images were acquired with a confocal laser scanning microscope (Fluoview ver. 3.1; Olympus, Japan).

P. Fang et al. / Virology 499 (2016) 170–177174

P. Fang et al. / Virology 499 (2016) 170–177 175

The NS6 protein was localized in the cytoplasm during the entireexperimental period (Fig. 5A).

To examine the precise subcellular localization of the NS6protein in PDCoV-infected LLC-PK cells, the cells were fixed at 12hpi and stained with both mouse anti-NS6 monoclonal antibody(MAb 4B9) and rabbit polyclonal antibodies against either theendoplasmic reticulum marker GRP94, the Golgi apparatus markerGS28, or the ER-Golgi intermediate compartment marker SEC31.NS6 protein was observed to predominantly colocalize with boththe ER compartment and the ER-Golgi intermediate compartment,as well as to partially colocalize with the Golgi complex (Fig. 5B).

3. Discussion

Although coronavirus accessory proteins are not essential forvirus replication in vitro (Casais et al., 2005; Yount et al., 2005),previous reports have made it clear that these accessory proteinsare not redundant, but rather possess many functions, includingmodulating viral pathogenicity (De Haan et al., 2002) and acting ascell death inducers (Law et al., 2005) or interferon antagonists(Niemeyer et al., 2013; Siu et al., 2014). Furthermore, recent studieshave reported the discovery of novel accessory proteins in somecoronaviruses, such as the newly identified accessory protein ORFXof bat severe acute respiratory syndrome-like coronavirus (Zenget al., 2016). For deltacoronavirus, no information on the expression,intracellular localization, function of accessory proteins or on theirunique sgRNAs has been reported. Indeed, given the absence of aTRS preceding NS6 and the relatively high (0.500) Ka/Ks ratio ofNS6 in Bulbul Coronavirus (BuCoV), another deltacoronavirus, itseems likely that BuCoV does not express this accessory protein(Woo et al., 2009). In contrast, a typical putative TRS preceding NS6is present in other deltacoronaviruses, such as PDCoV, White-eyeCoronavirus, and Sparrow Coronavirus (Woo et al., 2012). However,these deltacoronavirus accessory proteins have been predicted inthe absence of any experimental evidence to confirm their ex-pression during virus infection. Here, we identify the sgRNA of thePDCoV NS6 gene for the first time and provide experimental evi-dence to confirm its expression in PDCoV-infected cells.

To analyze the PDCoV sgRNAs, we initially used a non-radio-active labeling Northern blot. Unfortunately, NS6 sgRNA could notbe detected, even though other sgRNAs including S, M, and N,were easily identified (data not shown). This result may be due toa lower sensitivity of the non-radioactive labeling technique or alower abundance of NS6 sgRNA. Thus, we performed leader-bodyjunction RT-PCR, which had been extensively used in previousstudies (Hussain et al., 2005; Zhang et al., 2007), to examine theexistence of sgRNA NS6. The advantages of this method includehigher sensitivity and simplicity of operation. Moreover, it candetermine the leader-body junction site of each sgRNA by DNAsequencing. To avoid interference from other high abundancesgRNAs, such as sgRNAs S and N, the reverse primer was designedto target the end of NS6 and the extension time in each PCR cyclewas set as 20 s. As expected, only M and NS6 sgRNAs were am-plified in our experimental conditions (Fig. 1B).

Unexpectedly, the location and sequence of the leader-bodyjunction site of NS6 sgRNA did not match previous projections. Thepredicted leader-body fusion site for NS6 sgRNA is 46 nucleotidesupstream of the NS6 gene AUG start codon; however, sequenceanalysis of the RT-PCR product shows that the actual site for NS6sgRNA is located to 148 nucleotides upstream of the AUG start codonof NS6 gene. Additionally, each sgRNA contains a TRS characterizedby an AU-rich section of about 10 nucleotides (the core is a hex-anucleotide motif) (Bentley et al., 2013). The TRS is the fusion site ofleader and body sequences of coronavirus sgRNAs (Dye and Siddell,2005). Although the molecular mechanisms of the leader-body

fusion in coronavirus sgRNAs have not been elucidated fully, accu-mulating evidence suggests that most TRSs are derived from theleader TRS (TRS-L), which are considered canonical TRSs, while somenon-canonical TRSs are derived from the body TRS (TRS-B) (VanMarle et al., 1999; Zuniga et al., 2004). The conserved hexanucleotideTRS used by PDCoV is ACACCA, as determined for M sgRNA by ourstudy, while the TRS of NS6 sgRNA is ACACCT, which is identical tothe core sequence of TRS-B, suggesting that NS6 uses a non-canonicalTRS. Similar phenomena have also been found in sgRNAs 2-1 and 3-1of SARS-CoV (Hussain et al., 2005). The differences in TRS sequencefor different sgRNAs could play a regulatory role in controlling theabundances of different mRNAs (Bentley et al., 2013; Hussain et al.,2005; Zuniga et al., 2004). It is possible that the lower abundance ofNS6 sgRNA is associated with its non-canonical TRS.

Although many sgRNA of different coronaviruses have been iden-tified, their natural expression products in virus-infected cells are yetto be confirmed due to lack of specific antibodies. Here, we generatedtwo MAbs against PDCoV NS6 and demonstrated that the NS6 proteincould be detected in both PDCoV-infected cells and cells transfectedwith a eukaryotic expression construct of the NS6 gene. We also foundthat a similar intracellular localization distribution could be observedin both cells infected with PDCoV and those overexpressing NS6.Furthermore, the expression of NS6 could be detected at the earlyphase (6 hpi) of PDCoV infection. Subcellular localization analysesshowed that NS6 colocalized partially with the Golgi, and pre-dominantly with the ER complex and ER-Golgi intermediate com-partment. Similar subcellular localization has also been reported forSARS-CoV ORF7a (also known as U122), an accessory protein in-corporated into purified SARS-CoV particles (Fielding et al., 2004;Huang et al., 2006). It is well known that the ER-Golgi intermediatecompartment is the site of coronavirus assembly and packaging(Klumperman et al., 1994; McBride and Fielding, 2012). The presenceof NS6 in the ER-Golgi intermediate compartment suggests the pos-sibility that NS6 plays a role in viral assembly and budding events.Indeed, our preliminary study found that NS6 could be detected in thepurified virion by western blotting (data not shown); however, addi-tional experimental evidence, especially confirmation with im-munogold electron microscopy is required to support this idea.

Previous studies suggested that coronavirus accessory proteinsare often dispensable for virus replication in vitro, but are required foroptional replication and virulence in the natural host (McBride andFielding, 2012; Zhao et al., 2009). PDCoV infection suppresses RIG-I-mediated interferon-β production (Luo et al., 2016), and previousstudies suggested that some accessary proteins of coronavirus, suchas SARS-CoV ORF3b (Freundt et al., 2009) and MERS-CoV ORF4b(Thornbrough et al., 2016), are interferon antagonists. WhetherPDCoV NS6 possesses this property requires further study. In addi-tion, BLAST search revealed that NS6 proteins share no amino acidsimilarities with other CoV accessory proteins or known host pro-teins. No putative transmembrane domain and functional domainwere identified by TMHMM (www.cbs.dtu.dk/services/TMHMM) andInterProScan (www.ebi.ac.uk/interpro) analyses, respectively. Precisefunctions of NS6 protein need to be yet further explored. At present,we are making efforts to construct an infectious cDNA clone ofPDCoV and hope to use a reverse genetics system to study the NS6function in the PDCoV viral life cycle and pathogenesis.

In summary, we confirmed the existence of a separate NS6 sgRNAwith a non-canonical leader-body fusion site. We also prepared twoMAbs against PDCoV NS6 and demonstrated that NS6 proteins areindeed expressed in PDCoV-infected cells. The expression of NS6 couldbe detected as early as 6 hpi in the cytoplasm of PDCoV-infected cells.The expressed NS6 proteins in PDCoV-infected cells predominantlylocalize to the ER complex and ER-Golgi intermediate compartment.The identification of sgRNA NS6 and the expression and subcellularlocalization of NS6 protein lay the foundation for elucidation of thestructure and function of NS6 in the PDCoV viral replication cycle.

P. Fang et al. / Virology 499 (2016) 170–177176

4. Materials and methods

4.1. Cells, viruses, and reagents

LLC-PK and SP2/0 myeloma cells were respectively grown inDulbecco's Modified Eagle medium (DMEM) (Invitrogen) andRPMI 1640 Medium, each supplemented with 10% heat-inactivatedfetal bovine serum, at 37 °C in a humidified 5% CO2 incubator.PDCoV isolate CHN-HN-2014 (GenBank accession no. KT336560),which was isolated from a suckling piglet with acute diarrhea inChina in 2014, was propagated in LLC-PK cells in DMEM with10 μg/mL of trypsin. The titer of the supernatant from infectedcells was up to 108.5 TCID50 (median tissue culture infectious dose)per milliliter. Horseradish peroxidase-conjugated goat anti-mouseIgG, Alexa Fluor 594-conjugated donkey anti-mouse (rabbit) IgG,and Alexa Fluor 488-conjugated donkey anti-mouse (rabbit) IgGwere purchased from Santa Cruz Biotechnology, Inc. Mousemonoclonal antibodies against hemagglutinin were purchasedfrom Medical and Biological Laboratories (Japan).

4.2. Leader-body junction analysis

Total intracellular RNAwas extracted from PDCoV-infected LLC-PK cell lysates using Trizol reagent (Invitrogen) and then reverse-transcribed using a Transcription First Strand cDNA Synthesis kit(Roche) according to the manufacturer's instructions with specificprimer NS6r, which binds to a location within the C terminal of theNS6 gene. The obtained cDNAs were amplified by PCR with NS6rand a forward primer Leader-F, which binds to a location in the 5ʹ-leader sequence (Table 1). PCR reaction conditions were 94 °C for5 min, followed by 30 cycles of 94 °C for 30 s, 56 °C for 30 s, and72 °C for 20 s, and a final extension of 72 °C for 5 min. PCR pro-ducts were analyzed on a 1.2% agarose gel and then cloned into apMD18-T cloning vector (Takara) for sequencing.

4.3. Cloning of the NS6 gene

Viral genomic RNAwas extracted from the supernatant of PDCoV-infected LLC-PK cells using Trizol Reagent (Invitrogen) and usedimmediately for cDNA synthesis according to the manufacturer'sinstructions. NS6 gene-specific primers were designed using Primer5.0 software based on the published CHN-HN-2014 sequence (Ta-ble 1). The 285-bp fragment of the NS6 gene was amplified by RT-PCR and then cloned into a pET28a (þ) expression vector. The suc-cessful construction of the recombinant plasmid, named pET28a-NS6, was verified using enzyme analysis and DNA sequencing.

4.4. Preparation of the NS6 protein

The recombinant plasmid pET28a-NS6 and empty vector pET28a(þ) were separately transformed into Rosetta (DE3) cells and inducedby 0.8 mM IPTG for 3–6 h. The bacteria were then collected and wa-shed with phosphate-buffered saline (PBS), followed by cell lysis viathe supersonic schizolysis. SDS-PAGE was performed to analyze theNS6 protein expression. The proteins were purified by the supersonicschizolysis method as described previously (Liu et al., 2009), and theconcentration of purified proteins was subsequently determined byusing a Trace ultraviolet spectrophotometer.

4.5. Immunization of mice

Three 6-week-old female BALB/c mice were inoculated viasubcutaneous injection with purified NS6 protein (120 μg/mouse)mixed with the same amount of Freund's complete adjuvant. Twoweeks later, these mice were given a second immunization withthe NS6 protein (100 μg/mouse) emulsified in an equal amount of

Freund's incomplete adjuvant. Two weeks after the second in-oculation, mouse antisera were collected, and the titers of anti-body in the sera were assessed by performing indirect ELISA.Subsequently, mice with higher antibody titers were given abooster immunization of NS6 protein (200 μg) by intraperitonealinjection 3–5 days prior to their use in hybridoma production.

4.6. Production of anti-NS6 protein MAb

Splenocytes from the mouse with the highest antibody titer fol-lowing immunizationwith the NS6 proteinwere harvested and fusedwith SP2/0 myeloma cells under the condition of 50% polyethyleneglycol (PEG4000) as a fusion agent. The resulting hybridoma cellswere then cultured in 96-well plates at 37 °C in a humidified 5% CO2

incubator in HAT screening culture mediumwith fetal bovine serum.One week later, positive hybridomas with coverage of one third toone half of the bottom of 96-well plates were filtered by indirectELISA. Subsequently, the positive hybridoma cells were subclonedthree times by the limiting dilution method. Finally, the stable hy-bridoma cells were passaged, and the resulting cell supernatantswere collected for use in later experiments. Meanwhile, the stablehybridoma cells were stored in liquid nitrogen.

4.7. Indirect immunofluorescence assays (IFAs) and confocalmicroscopy

LLC-PK cells in 24-well plates with 80% confluence were trans-fected with the plasmid encoding the putative NS6 gene or an emptyvector for 28 h. Separately, similar monolayers of LLC-PK cells in 24-well plates were mock-infected or infected with PDCoV and thencultured in the DMEM containing trypsin (10 μg/mL) at 37 °C. Bothsets of cells were then fixed, followed by use in indirect IFAs with thefollowing primary antibodies: NS6-specific MAbs (4B9) and rabbitanti-HSP90B1 (GRP94, ER marker; ABclonal) polyclonal antibody,rabbit anti-GOSR1 (GS28, Golgi marker; ABclonal) polyclonal antibody,or rabbit anti-SEC31A (SEC31, ER-Golgi intermediate compartmentmarker; ABclonal) polyclonal antibody (1:100 dilution in PBS). AlexaFluor 594-conjugated donkey anti-mouse (rabbit) IgG or Alexa Fluor488-conjugated donkey anti-mouse (rabbit) IgG (1:500 dilution inPBST) was then used as secondary antibodies. Cells were treated with4,6-diamidino-2-phenylindole (DAPI) to indicate the locations of thecell nuclei. Fluorescent images were obtained with a confocal laserscanning microscope (Fluoview ver. 3.1; Olympus, Japan).

4.8. Western blotting analysis

For the western blotting analysis, cells that had been transfected orinfected as described above were lysed, collected, and separated bySDS-PAGE with 15% polyacrylamide gels. Subsequently, the separatedprotein was transferred to a nitrocellulose membrane. The membranewas blocked with 5% skim milk in PBST with 0.1% polysorbate-20.NS6-specific MAbs were then incubated with the nitrocellulosemembrane for 3 h at room temperature. After washing three timeswith PBST, themembranewas incubatedwith horseradish peroxidase-conjugated goat anti-mouse IgG (1:5000 dilution, in PBST) for 45 minat room temperature. After washing three times, the membrane wasvisualized by enhanced chemiluminescence reagents (ELC; BIO-RAD).

Acknowledgements

This work was supported by the National Key R&D Plan ofChina (2016YFD0500103), the Key Technology R&D Programme ofChina (2015BAD12B02), and the Natural Science Foundation ofHubei Province (2014CFA009).

P. Fang et al. / Virology 499 (2016) 170–177 177

References

Bentley, K., Keep, S.M., Armesto, M., Britton, P., 2013. Identification of a non-canonically transcribed subgenomic mRNA of infectious bronchitis virus andother gammacoronaviruses. J. Virol. 87, 2128–2136.

Casais, R., Davies, M., Cavanagh, D., Britton, P., 2005. Gene 5 of the avian coronavirusinfectious bronchitis virus is not essential for replication. J. Virol. 79,8065–8078.

Chen, F., Zhu, Y., Wu, M., Ku, X., Yao, L., He, Q., 2015a. Full-length genome char-acterization of chinese porcine deltacoronavirus strain CH/SXD1/2015. GenomeAnnounc. 3, e01284-15.

Chen, Q., Gauger, P., Stafne, M., Thomas, J., Arruda, P., Burrough, E., Madson, D.,Brodie, J., Magstadt, D., Derscheid, R., Welch, M., Zhang, J., 2015b. Pathogenicityand pathogenesis of a United States porcine deltacoronavirus cell culture iso-late in 5-day-old neonatal piglets. Virology 482, 51–59.

De Haan, C.A., Masters, P.S., Shen, X., Weiss, S., Rottier, P.J., 2002. The group-specificmurine coronavirus genes are not essential, but their deletion, by reverse ge-netics, is attenuating in the natural host. Virology 296, 177–189.

Dong, N., Fang, L.R., Zeng, S.L., Sun, Q.Q., Chen, H.C., Xiao, S.B., 2015. Porcine del-tacoronavirus in mainland china. Emerg. Infect. Dis. 21, 2254–2255.

Dye, C., Siddell, S.G., 2005. Genomic RNA sequence of Feline coronavirus strain FIPVWSU-79/1146. J. Gen. Virol. 86, 2249–2253.

Fielding, B.C., Tan, Y.J., Shuo, S., Tan, T.H., Ooi, E.E., Lim, S.G., Hong, W., Goh, P.Y.,2004. Characterization of a unique group-specific protein (U122) of the severeacute respiratory syndrome coronavirus. J. Virol. 78, 7311–7318.

Freundt, E.C., Yu, L., Park, E., Lenardo, M.J., Xu, X.N., 2009. Molecular determinantsfor subcellular localization of the severe acute respiratory syndrome cor-onavirus open reading frame 3b protein. J. Virol. 83, 6631–6640.

Haijema, B.J., Volders, H., Rottier, P.J., 2004. Live, attenuated coronavirus vaccinesthrough the directed deletion of group-specific genes provide protectionagainst feline infectious peritonitis. J. Virol. 78, 3863–3871.

Hu, H., Jung, K., Vlasova, A.N., Chepngeno, J., Lu, Z.Y., Wang, Q.H., Saif, L.J., 2015.Isolation and characterization of porcine deltacoronavirus from pigs withdiarrhea in the United States. J. Clin. Microbiol. 53, 1537–1548.

Huang, C., Ito, N., Tseng, C.T.K., Makino, S., 2006. Severe acute respiratory syndromecoronavirus 7a accessory protein is a viral structural protein. J. Virol. 80,7287–7294.

Hussain, S., Pan, J., Chen, Y., Yang, Y.L., Xu, J., Peng, Y., Wu, Y., Li, Z.Y., Zhu, Y., Tien, P.,Guo, D.Y., 2005. Identification of novel subgenomic RNAs and noncanonicaltranscription initiation signals of severe acute respiratory syndrome cor-onavirus. J. Virol. 79, 5288–5295.

Janetanakit, T., Lumyai, M., Bunpapong, N., Boonyapisitsopa, S., Chaiyawong, S.,Nonthabenjawan, N., Kesdaengsakonwut, S., Amonsin, A., 2016. Porcine delta-coronavirus, Thailand, 2015. Emerg. Infect. Dis. 22, 757–759.

Jung, K., Hu, H., Saif, L.J., 2016. Porcine deltacoronavirus infection: etiology, cellculture for virus isolation and propagation, molecular epidemiology and pa-thogenesis. Virus Res. . http://dx.doi.org/10.1016/j.virusers. 2016.04.009

Jung, K., Hu, H., Eyerly, B., Lu, Z., Chepngeno, J., Saif, L.J., 2015. Pathogenicity of2 porcine deltacoronavirus strains in gnotobiotic pigs. Emerg. Infect. Dis. 21,650–654.

Klumperman, J., Locker, J.K., Meijer, A., Horzinek, M.C., Geuze, H.J., Rottier, P.J., 1994.Coronavirus M proteins accumulate in the Golgi complex beyond the site ofvirion budding. J. Virol. 68, 6523–6534.

Kopecky-Bromberg, S.A., Martinez-Sobrido, L., Frieman, M., Baric, R.A., Palese, P.,2007. Severe acute respiratory syndrome coronavirus open reading frame (ORF)3b, ORF 6, and nucleocapsid proteins function as interferon antagonists. J. Virol.81, 548–557.

Law, P.T., Wong, C.H., Au, T.C., Chuck, C.P., Kong, S.K., Chan, P.K., To, K.F., Lo, A.W.,Chan, J.Y., Suen, Y.K., Chan, H.Y., Fung, K.P., Waye, M.M., Sung, J.J., Lo, Y.M., Tsui,S.K., 2005. The 3a protein of severe acute respiratory syndrome-associatedcoronavirus induces apoptosis in Vero E6 cells. J. Gen. Virol. 86, 1921–1930.

Lee, J.H., Chung, H.C., Nguyen, V.G., Moon, H.J., Kim, H.K., Park, S.J., Lee, C.H., Lee, G.E., Park, B.K., 2016. Detection and phylogenetic analysis of porcine deltacor-onavirus in Korean swine farms, 2015. Transbound. Emerg. Dis. 63, 248–252.

Lee, S., Lee, C., 2014. Complete genome characterization of Korean porcine delta-coronavirus strain KOR/KNU14-04/2014. Genome Announc. 2, e01191-14.

Lee, S., Lee, C., 2015. Functional characterization and proteomic analysis of thenucleocapsid protein of porcine deltacoronavirus. Virus Res. 208, 136–145.

Li, G., Chen, Q., Harmon, K.M., Yoon, K.J., Schwartz, K.J., Hoogland, M.J., Gauger, P.C.,Main, R.G., Zhang, J., 2014. Full-length genome sequence of porcine deltacor-onavirus strain USA/IA/2014/8734. Genome Announc. 2, e00278-14.

Liu, B., Li, G., Sui, X., Yin, J., Wang, H., Ren, X., 2009. Expression and functionalanalysis of porcine aminopeptidase N produced in prokaryotic expressionsystem. J. Biotechnol. 141, 91–96.

Liu, D.X., Fung, T.S., Chong, K.K., Shukla, A., Hilgenfeld, R., 2014. Accessory proteinsof SARS-CoV and other coronaviruses. Antiviral Res. 109, 97–109.

Lorsirigool, A., Saeng-Chuto, K., Temeeyasen, G., Madapong, A., Tripipat, T., Wegner,M., Tuntituvanont, A., Intrakamhaeng, M., Nilubol, D., 2016. The first detectionand full-length genome sequence of porcine deltacoronavirus isolated in LaoPDR. Arch. Virol. 161, 2909–2911.

Luo, J., Fang, L., Dong, N., Fang, P., Ding, Z., Wang, D., Chen, H., Xiao, S., 2016. Porcinedeltacoronavirus (PDCoV) infection suppresses RIG-I-mediated interferon-βproduction. Virology 495, 10–17.

Ma, Y., Zhang, Y., Liang, X., Lou, F., Oglesbee, M., Krakowka, S., Li, J., 2015. Origin,evolution, and virulence of porcine deltacoronaviruses in the United States.MBio 6, e00064.

Madapong, A., Saeng-Chuto, K., Lorsirigool, A., Temeeyasen, G., Srijangwad, A.,Tripipat, T., Wegner, M., Nilubol, D., 2016. Complete genome sequence of por-cine deltacoronavirus isolated in Thailand in 2015. Genome Announc. 4,e00408–e00416.

Marthaler, D., Jiang, Y., Collins, J., Rossow, K., 2014a. Complete genome sequence ofstrain SDCV/USA/Illinois121/2014, a porcine deltacoronavirus from the UnitedStates. Genome Announc. 2, e00218-14.

Marthaler, D., Raymond, L., Jiang, Y., Collins, J., Rossow, K., Rovira, A., 2014b. Rapiddetection, complete genome sequencing, and phylogenetic analysis of porcinedeltacoronavirus. Emerg. Infect. Dis. 20, 1347–1350.

McBride, R., Fielding, B.C., 2012. The role of severe acute respiratory syndrome(SARS)-coronavirus accessory proteins in virus pathogenesis. Viruses 4,2902–2923.

Niemeyer, D., Zillinger, T., Muth, D., Zielecki, F., Horvath, G., Suliman, T., Barchet, W.,Weber, F., Drosten, C., Muller, M.A., 2013. Middle east respiratory syndromecoronavirus accessory protein 4a is a type I interferon antagonist. J. Virol. 87,12489–12495.

Sawicki, D., Wang, T., Sawicki, S., 2001. The RNA structures engaged in replicationand transcription of the A59 strain of mouse hepatitis virus. J. Gen. Virol. 82,385–396.

Sawicki, S.G., Sawicki, D.L., Siddell, S.G., 2007. A contemporary view of coronavirustranscription. J. Virol. 81, 20–29.

Siu, K.L., Yeung, M.L., Kok, K.H., Yuen, K.S., Kew, C., Lui, P.Y., Chan, C.P., Tse, H., Woo,P.C., Yuen, K.Y., Jin, D.Y., 2014. Middle east respiratory syndrome coronavirus 4aprotein is a double-stranded RNA-binding protein that suppresses PACT-in-duced activation of RIG-I and MDA5 in the innate antiviral response. J. Virol. 88,4866–4876.

Song, D., Zhou, X., Peng, Q., Chen, Y., Zhang, F., Huang, T., Zhang, T., Li, A., Huang, D.,Wu, Q., He, H., Tang, Y., 2015. Newly emerged porcine deltacoronavirus asso-ciated with diarrhoea in swine in china: identification, prevalence and full-length genome sequence analysis. Transbound. Emerg. Dis. 62, 575–580.

Tan, Y.J., Lim, S.G., Hong, W., 2006. Understanding the accessory viral proteinsunique to the severe acute respiratory syndrome (SARS) coronavirus. AntiviralRes. 72, 78–88.

Thornbrough, J.M., Jha, B.K., Yount, B., Goldstein, S.A., Li, Y., Elliott, R., Sims, A.C.,Baric, R.S., Silverman, R.H., Weiss, S.R., 2016. Middle east respiratory syndromecoronavirus NS4b protein inhibits host RNase L activation. MBio 7, e00258.

Van Marle, G., Dobbe, J.C., Gultyaev, A.P., Luytjes, W., Spaan, W.J., Snijder, E.J., 1999.Arterivirus discontinuous mRNA transcription is guided by base pairing be-tween sense and antisense transcription-regulating sequences. Proc. Natl. Acad.Sci. USA 96, 12056–12061.

Wang, L., Byrum, B., Zhang, Y., 2014a. Detection and genetic characterization ofdeltacoronavirus in pigs, Ohio, USA, 2014. Emerg. Infect. Dis. 20, 1227–1230.

Wang, L., Byrum, B., Zhang, Y., 2014b. Porcine coronavirus HKU15 detected in 9 USstates, 2014. Emerg. Infect. Dis. 20, 1594–1595.

Wang, Y.W., Yue, H., Fang, W., Huang, Y.W., 2015. Complete genome sequence ofporcine deltacoronavirus strain CH/Sichuan/S27/2012 from mainland china.Genome Announc. 3, e00945-15.

Woo, P.C.Y., Huang, Y., Lau, S.K.P., Yuen, K.Y., 2010. Coronavirus genomics andbioinformatics analysis. Viruses 2, 1804–1820.

Woo, P.C.Y., Lau, S.K.P., Lam, C.S.F., Lai, K.K.Y., Huang, Y., Lee, P., Luk, G.S.M., Dyrting,K.C., Chan, K.H., Yuen, K.Y., 2009. Comparative analysis of complete genomesequences of three avian coronaviruses reveals a novel group 3c coronavirus. J.Virol. 83, 908–917.

Woo, P.C.Y., Lau, S.K.P., Lam, C.S.F., Lau, C.C.Y., Tsang, A.K.L., Lau, J.H.N., Bai, R., Teng, J.L.L., Tsang, C.C.C., Wang, M., Zheng, B.J., Chan, K.H., Yuen, K.Y., 2012. Discovery ofseven novel Mammalian and avian coronaviruses in the genus deltacoronavirussupports bat coronaviruses as the gene source of alphacoronavirus and beta-coronavirus and avian coronaviruses as the gene source of gammacoronavirusand deltacoronavirus. J. Virol. 86, 3995–4008.

Yount, B., Roberts, R.S., Sims, A.C., Deming, D., Frieman, M.B., Sparks, J., Denison, M.R., Davis, N., Baric, R.S., 2005. Severe acute respiratory syndrome coronavirusgroup-specific open reading frames encode nonessential functions for re-plication in cell cultures and mice. J. Virol. 79, 14909–14922.

Zeng, L.P., Gao, Y.T., Ge, X.Y., Zhang, Q., Peng, C., Yang, X.L., Tan, B., Chen, J., Chmura,A.A., Daszak, P., Shi, Z.L., 2016. Bat severe acute respiratory syndrome-likecoronavirus WIV1 encodes an extra accessory protein, ORFX, involved inmodulation of the host immune response. J. Virol. 90, 6573–6582.

Zhang, J., 2016. Porcine deltacoronavirus: overview of infection dynamics, diag-nostic methods, prevalence and genetic evolution. Virus Res. . http://dx.doi.org/10.1016/j.virusers.2016.05.028

Zhang, J., Guy, J.S., Snijder, E.J., Denniston, D.A., Timoney, P.J., Balasuriya, U.B., 2007.Genomic characterization of equine coronavirus. Virology 369, 92–104.

Zhao, J., Falcon, A., Zhou, H., Netland, J., Enjuanes, L., Perez Brena, P., Perlman, S.,2009. Severe acute respiratory syndrome coronavirus protein 6 is required foroptimal replication. J. Virol. 83, 2368–2373.

Zuniga, S., Sola, I., Alonso, S., Enjuanes, L., 2004. Sequence motifs involved in theregulation of discontinuous coronavirus subgenomic RNA synthesis. J. Virol. 78,980–994.