key: cord-340438-9q3ic0ye
authors: Zhang, Jianqiang; Yim-Im, Wannarat; Chen, Qi; Zheng, Ying; Schumacher, Loni; Huang, Haiyan; Gauger, Phillip; Harmon, Karen; Li, Ganwu
title: Identification of porcine epidemic diarrhea virus variant with a large spike gene deletion from a clinical swine sample in the United States
date: 2018-02-21
journal: Virus Genes
DOI: 10.1007/s11262-018-1542-7
sha: 
doc_id: 340438
cord_uid: 9q3ic0ye

Two genetically different porcine epidemic diarrhea virus (PEDV) strains have been identified in the USA: US prototype (also called non-S INDEL) and S INDEL PEDVs. In February 2017, a PEDV variant (USA/OK10240-8/2017) was identified in a rectal swab from a sow farm in Oklahoma, USA. Complete genome sequence analyses indicated this PEDV variant was genetically similar to US non-S INDEL strain but had a continuous 600-nt (200-aa) deletion in the N-terminal domain of the spike gene compared to non-S INDEL PEDVs. This is the first report of detecting PEDV bearing large spike gene deletion in clinical swine samples in the USA.

Porcine epidemic diarrhea virus (PEDV) is the causative agent of porcine epidemic diarrhea (PED) that was first recorded in Europe in the 1970s [1, 2] . PEDV spread to Asia during the 1980s and 1990s and became endemic in pigs in Asian countries [3] . In 2010, a severe PED outbreak occurred in China characterized by high morbidity in pigs of all ages and high mortality in neonatal piglets [4, 5] . In 2013, PED outbreaks were reported for the first time in the USA [6] and caused substantial economic losses [7] . Subsequently, US-like PEDVs were identified in other American countries and also emerged or re-emerged in some Asian and European countries [8] . Global PEDVs exhibit significant genetic diversities. Recently, Lin et al. [8] proposed to categorize global PEDV strains into Classical, S INDEL, emerging North American non-S INDEL, and emerging Asian non-S INDEL strains. In the USA, at least two genetically different PEDV strains have been identified: the highly virulent PEDV first identified in April 2013 associated with severe PED outbreaks was referred to as 'US prototype' or 'US original' or 'non-S INDEL' strain [9, 10] ; a clinically milder PEDV variant identified in the USA in January 2014 which was different from the original highly virulent PEDV strains, as reflected by insertions and deletions in the spike (S) gene, was designated as 'S INDEL' PEDV [10, 11] . In this case report, we describe, for the first time, identification of a PEDV variant with a large spike gene N-terminal domain deletion from a clinical swine sample in the USA.

At the Iowa State University Veterinary Diagnostic Laboratory (ISU VDL), a nucleocapsid (N) gene-based real-time RT-PCR (rRT-PCR) is routinely used for the screening detection of PEDV from clinical specimens [12] [13] [14] . If positive, a spike gene-based multiplex rRT-PCR can be further used to differentiate non-S INDEL from S INDEL PEDV strains. In February 2017, rectal swabs collected from a Edited by Juergen A Richt.

GenBank Accession Numbers: The complete genome sequences of two porcine epidemic diarrhea viruses described in this study have been deposited in GenBank under accessions MG334554 and MG334555. sow farm in Oklahoma, USA, were submitted to the ISU VDL for PEDV PCR testing. The samples were positive for PEDV by the N gene-based rRT-PCR. Subsequent PEDV S gene-based differential rRT-PCR revealed that these samples were negative for S INDEL PEDV (C T > 40) but positive for non-S INDEL PEDV. Generally, the PEDV S gene-based differential rRT-PCR gave 2-3 C T higher than the N genebased rRT-PCR on the same sample. However, the sample #8 gave unexpected results: strong positive by N gene-based rRT-PCR (C T 15.5) but weak positive for non-S INDEL PEDV by the differential rRT-PCR (C T 36.8). To determine the possible reasons for this observation, the sample #8 and another control sample #6 (C T 18.2 by N gene-based rRT-PCR and C T 20.8 for non-S INDEL by the S gene-based differential rRT-PCR) were sequenced using next-generation sequencing technology following previously described procedures [15, 16] . The PEDV in the sample #6 (USA/ OK10240-6/2017) and the sample #8 (USA/OK10240-8/2017) had whole genome sequences of 28,038 and 27,438 nucleotides in length, respectively. The sequences of these two PEDVs have been deposited into GenBank (MG334554 and MG334555). Phylogenetic analyses based on the whole genome sequences and the spike gene indicated that both OK10240-6 and OK10240-8 belong to the US non-S INDEL cluster (Fig. 1) . However, compared to the OK10240-6 and other non-S INDEL PEDV strains, the OK10240-8 PEDV had a large continuous deletion of 600-nt (200-aa) in the spike gene/protein (nt ∆91-690; aa ∆31-230; Fig. 2 ). The remaining genome of the OK10240-8 PEDV, other than the S deletion region, had approximately 99.7% nt identity to other non-S INDEL PEDV strains. A gel-based RT-PCR [17] was used to differentiate the OK10240-8 PEDV from non-S INDEL PEDV. Twenty more samples were collected from the same farm; all of them contained non-S INDEL PEDV but none of them contained OK10240-8-like PEDV, indicating the prevalence of OK10240-8-like PEDV in swine populations may be very low. Virus isolation attempts on the sample #8 in Vero cells (ATCC CCL-81) were unsuccessful. The remaining sample #8 (250 µl diluted in 2250 µl culture medium) was orally inoculated into two 10-day-old PEDV-negative piglets (10 ml/pig) but did not result in active infection.

PEDV spike (S) protein is a type I membrane glycoprotein with a signal peptide (amino acid residues 1-18), a large [18] . The S protein assembles into homotrimers that form the clubshaped projections (spikes) on the virion surface. PEDV S protein has multiple functions including (1) mediating receptor binding through its S1 subunit (aa 1-729) and fusion of the viral and cellular membranes during cell entry through its S2 subunit (aa 730-1386); (2) harboring neutralization epitopes. Specifically, the N-terminal domain (aa 19-233) exhibits sialic acid binding activity; the receptor-binding domain (aa 501-629) is believed to interact with a protein receptor; and a fusion peptide domain (aa 891-908) mediates virus-cell membrane fusion during cell entry [18] . Neutralization epitopes have been reported within the amino acid residues 1-219, 499-638, 636-789, and 1371-1377 [19] [20] [21] [22] . The aminopeptidase N protein (APN) serves as a receptor for several alphacoronaviruses such as canine coronavirus type II, feline coronavirus type II, transmissible gastroenteritis virus (TGEV), porcine respiratory coronavirus (PRCV), and human coronavirus 229E [18] . Porcine APN was considered to be the putative receptor of PEDV with some supporting evidence [23] [24] [25] [26] ; however, some recent studies indicate that porcine APN may not be a functional receptor for PEDV [27, 28] .

The N-terminal domain of PEDV S protein is one of the most variable regions in the PEDV genome. The insertions and deletions of S INDEL PEDV strains and the large deletion (aa ∆31-230) of the PEDV variant (OK10240-8) identified in this study are all located in the N-terminal domain region. It is predicted that deletion of 200-aa at this region would not interfere with either the protein receptor binding or the neutralization epitopes 499-638, 636-789, and 1371-1377. 3-D structural analyses of the S protein also suggest that this 200-aa deletion may not interfere with trimer formation. However, the neutralization epitope within residues 1-129 and the sialic acid binding activity of the virus may be affected by this 200aa deletion. In fact, some studies have shown that PEDV strains having variations in the N-terminal domain of S protein exhibited different sialic acid binding activities [18, 29] . It remains to be determined whether the activity of sialic acid binding by the S protein affects virus entry into cells, replication in cells, and pathogenicity in pigs. In addition, 200-aa deletion in the OK10240-8 PEDV variant may affect virus virulence and pathogenicity. Construction of a recombinant PEDV carrying 200-aa deletion in this It was previously reported [30] that a cell cultureadapted US PEDV isolate TC-PC177-P2 contained 591nt (197-aa) deletion in the S protein (aa ∆34-230) but such deletions were not present in the original clinical sample OH/PC177/2013 (Fig. 2) . A Japanese PEDV strain Tottori2/2014, identified in a clinical sample, contained 582-nt (194-aa) deletion in the S protein (aa ∆23-216) [31] . A Korean PEDV strain MF3809/2008, identified in a clinical sample, contained a 612-nt (204-aa) deletion in the S protein but in a different location (aa ∆713-916) [32] . A recent study reported the coexistence of PEDV with a large S gene deletion and PEDV with intact S gene in domestic pigs in Japan [33] . It has been demonstrated that the USA/ TC-PC177-P2 and JPN/Tottori2/2014 isolates harboring a large S gene deletion are less virulent than non-S INDEL PEDVs in experimentally inoculated pigs [8, 29, 34] . In terms of TGEV, a large (224-aa) deletion in the spike gene changed the viral tropism from intestinal to respiratory and this TGEV mutant was later renamed as PRCV [35] . In contrast, the PEDV variant TC-PC177-P2 with large S gene deletion did not change intestinal tropism [8, 29] .

In summary, a new PEDV variant strain (USA/ OK10240-8/2017) belonging to the non-S INDEL cluster but with a 600-nt deletion (200-aa deletion) in the N-terminal domain of the S gene was identified in this study. This is the first report of a PEDV strain with a large deletion in the S gene identified in clinical swine samples in the USA. This PEDV with large S gene deletion was present on the same farm where non-S INDEL PEDV with intact S gene was detected but it appeared that the prevalence of OK10240-8-like PEDV in swine populations may be low. Additional molecular epidemiological studies are needed to monitor the emergence of novel PEDV variants and determine their prevalence levels in US swine.

Pig Farm

Acknowledgement This study was supported by the Iowa State Uni-