First detection of porcine respirovirus 1 in Germany and the Netherlands

Abstract Porcine respirovirus 1, also referred to as porcine parainfluenza virus 1 (PPIV‐1), was first detected in deceased pigs from Hong Kong in 2013. It has since then been found in the USA, Chile and most recently in Hungary. Information on the pathogenicity and global spread is sparse. However, it has been speculated to play a role in the porcine respiratory disease complex. To investigate the porcine virome, we screened 53 pig samples from 26 farms within the Dutch–German border region using shotgun metagenomics sequencing (SMg). After detecting PPIV‐1 in five farms through SMg, a real‐time reverse transcriptase PCR (RT‐qPCR) assay was designed, which not only confirmed the presence of the virus in 1 of the 5 farms but found an additional 6 positive farms. Phylogenetic analysis found the closest match to be the first detected PPIV‐1 strain in Hong Kong. The Dutch‐German region represents a significant area of pig farming within Europe and could provide important information on the characterization and circulation of porcine viruses, such as PPIV‐1. With its recent detection in Hungary, these findings suggest widespread circulation of PPIV‐1 in Central Europe, highlighting the need for further research on persistence, pathogenicity and transmission in Europe.


| INTRODUC TI ON
Viruses that belong to the Paramyxoviridae family have been associated with respiratory symptoms in farm animals and may cross host species barriers (Welch et al., 2017). Porcine respirovirus 1, also referred to as porcine parainfluenza virus 1 (PPIV-1), belongs to the Respirovirus genus within the Paramyxoviridae family. While the pathogenicity of PPIV-1 in pigs is still unclear, it has been isolated from pigs with respiratory symptoms (Lau et al., 2013). Challenge trials that involved inoculating conventional and CD/CD piglets with PPIV-1 revealed minimal morbidity and no mortality, despite considerable viral replication occurring in the infected animals ). Yet, PPIV-1 has been detected in co-infections with swine influenza virus (SIV) and porcine reproductive and respiratory syndrome virus (PRRSV), which suggests it could also play a role within the porcine respiratory disease complex (Welch et al., 2017). Information on the global spread and epidemiology is sparse. Apart from China (Lau et al., 2013), PPIV-1 was only detected in the United States of America (USA) in 2016 (Palinski et al., 2016), in Chile in 2020 (Agüero et al., 2020) and in Hungary also in 2020 (Dénes et al., 2020).
The Dutch-German region represents a significant area of pig farming within Europe and could provide crucial information on the circulation of porcine viruses. The Food Pro-tec-ts project (Food protection technologies for trans-boundary systems, work package TIC 2; (https://www.foodp rotec ts.eu/proje kt/arbei tspak ete/tic2/) aims to improve the early warning strategies for infectious diseases. For this purpose, we evaluated the use of shotgun metagenomics (SMg) as a method for unbiased detection of pathogens, including viruses, in a collection of nasal swab (NS), blood serum (BS) and oral fluid (OF) samples from pigs within the Dutch-German border region. After detecting PPIV-1 in five farms through SMg, a real-time reverse transcriptase PCR (RT-qPCR) assay was designed and 13 other farms were tested.

| Sample collection and selection
Samples were collected between June 2017 and October 2018 from farms within the Dutch-German border region under the Food Protects project. NS, OF and BS samples were collected from pigs in farms that monitored SIV, PRRSV and Salmonella spp. The NS and BS samples consisted of 5 pooled individual specimens that were obtained from pigs from the same age group and from the same herd. The OF samples were pen-based (a rope was used to collect OF specimens from all animals within a pen). All samples were frozen at −20°C after collection, shipped on ice and then stored at −80°C.

| Next-generation sequencing
Samples were initially chosen for SMg based on positive RT-qPCR results for SIV (VetMAX™ Gold SIV Detection Kit, Life Technologies) and/or PRRSV (Virotype, Hilden, Qiagen), as they are clinically relevant porcine pathogens. Overall, 34 BS (from 10 farms), 4 NS (from 4 farms) and 15 OF samples (from 12 farms) were sequenced using SMg, resulting in a total of 53 samples from 26 farms. Nucleic acids were extracted using the QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany), including in-column DNase digestion. Complementary DNA was synthesized, as described previously (Kafetzopoulou et al., 2018). Short-read sequencing (SRS) libraries were generated using the KAPA HyperPlus Kit (Roche, Basel, Switzerland).
Additionally, viral enrichment with oligonucleotide bait probes was performed on a selected sample using the SeqCap ViroCap share developer panel (Wylie et al., 2015). SRS libraries were se- To obtain rapid viral detection and taxonomic identification, raw reads were uploaded onto Taxonomer (IDbyDNA, San Carlos, CA, USA). Following the detection of PPIV-1, sequenced reads were first trimmed, then mapped against a PPIV-1 database derived from available complete or near-complete genomes on GenBank (n = 10, 27/08/2020) using CLC Genomics Workbench v.20.0.4 (CLC) (Qiagen, Hilden, Germany). The resulting consensus sequences had to cover >300 nucleotides (nt) of the respective reference genome to be considered valid and were subsequently confirmed and characterized by NCBI BLAST. Assembly was only performed on the sample BS-1 as it had a high abundance of PPIV-1 reads, using the best PPIV-1 hit as guidance on CLC.
The RT-qPCR was performed using the recommended settings for the Brilliant II RT-qPCR Master Mix 1-Step (Agilent Technologies, Texas, USA) with a MxPro30005P system (Agilent). Briefly, 400 nM of the forward and reverse primer and 200 nM of the probe were used to create a total reaction volume of 25 μL. The reaction was performed with the following cycling conditions: 50°C for 45 min, 95°C for 10 min, followed by 45 cycles of 95°C for 20 s, 55°C for 60 s and 72°C for 30 s. A cut-off value of Ct ≤40 was applied. The RT-qPCR was performed on samples that had sufficient sample material remaining following SMg. An additional 17 OF samples, either positive for SIV and/or PRRSV but not previously analysed with SMg, were also included. These 17 samples were collected from a further 13 different farms within the Dutch-German border region within the same time frame.

| Phylogenetic analysis
To create the PPIV-1 phylogenetic tree, the near-complete assembled PPIV-1 genome from this study and PPIV-1 genomes from GenBank (n = 10) were aligned with MAFFT v7.313 (Katoh et al., 2002). Phylogenetic trees were inferred from the alignments with the maximum likelihood approach implemented in RAxML v8.2.10 (Stamatakis et al., 2014) under the general time reversible (GTR-G) substitution model (Stamatakis et al., 2014) and bootstrap of 1,000 replicates. A phylogenetic analysis of the L (RdRp Polymerase) and the F (Fusion protein) genes was also performed.

| RE SULTS AND D ISCUSS I ON
The untargeted nature of SMg enables hypothesis-free approaches for exploration, surveillance and characterization of the porcine virome. Using this approach, we detected PPIV-1, an unexpected and rare virus, not previously seen in the Dutch-German border region, representing a major pig farming area in Europe (Figure 1a).
Additionally, we designed a specific RT-qPCR to test all available samples (both previously tested with and without SMg) to screen for PPIV-1 in other farms.
Using the SMg approach, PPIV-1 was detected in 5/53 samples (9.43%) from 3 different German farms and 2 different Dutch farms (5/26 farms; 19.2%) ( Table 1) (Table 1). One sample (OF-2) was positive using the SMg and RT-qPCR approaches. Sample BS-1 did F I G U R E 1 (a) European countries with reported PPIV-1 detection. Hungary (green) in 2020, Germany and the Netherlands (blue) in 2020. (b) Relative geographic location of the pre-selected German and Dutch farms. Farms with a PPIV-1 detection are illustrated with a red square, and farms without a PPIV-1 detection are illustrated with a blue hexagon. (c) Phylogenetic analysis of genome sequences of PPIV-1. The phylogenetic tree was constructed using the sequence obtained in this study (blue) and all complete or near-complete PPIV-1 genomes available in GenBank (n = 10, 27/08/2020). Human parainfluenza virus 1 was used as an outgroup have a Ct value of 44, and however, this was above the RT-qPCR cut-off (Ct ≤40). The design of specific PPIV-1 primers proved to be challenging due to the limited amount of available PPIV-1 sequences on GenBank. Although the RT-qPCR assay did reveal positive detection in samples not previously tested by SMg, it could not confirm PPIV-1 detection in all SMg-positive samples. A recently published paper designed a PPIV-1 RT-qPCR assay with primers based on the NCBI GenBank sequences, but contrary to our design, targeted the hemagglutinin-neuraminidase gene (Li et al., 2021). This RT-qPCR approach should be able to detect the strains described in our study (in silico prediction). In the future, it could be applied to screen for PPIV-1 and to pre-select samples for NGS to obtain further complete or near-complete sequences, not only to refine RT-qPCR assays but also to ensure optimal surveillance. Overall, PPIV-1 was detected in 1/34 BS samples (covering 10 farms), 2/4 NS samples (covering 4 farms) and 8/32 OF samples (covering 21 farms). In total, it was detected in 11 herds (31.4%) from 7 German farms and 4 Dutch farms by SMg and/or RT-qPCR (Table 1).
PPIV-1 genome coverage obtained using SMg ranged from 5.5% to 99.7% (Table 2). The sequencing of host, environmental and nonpathogenic nucleic acids, along with sequences of interest, reduces the sensitivity of SMg (Greninger, 2018). In this study, SMg did not detect viruses with Ct above 30. Therefore, viral enrichment strategies to obtain near-or complete viral sequences have been applied previously, including cell culture (Lau et al., 2013;Palinski et al., 2016) followed by depletion of ribosomal background RNA (Agüero et al., 2020).

RT-qPCR Ct value
Country of origin Application of oligonucleotide capture probes for viral enrichment on sample BS-1, which had the highest number of reads, resulted in a 22.8-fold increase in PPIV-1 reads and enabled the assembly of a near-complete genome sequence (15,344 nt) (GenBank accession: MT995732). This corresponded to 99.7% of the genome with a sequencing depth of 9,793 times. Furthermore, within BS-1 we were also able to identify a co-infection with PRRSV. Co-infections with PRRSV and/or SIV have been reported previously (Lau et al., 2013;Welch et al., 2017), with PPIV-1 being speculated to play a synergistic role within the porcine respiratory disease complex. However, the role of these co-infections in the porcine respiratory disease complex remains to be ascertained (Park et al., 2019;Welch et al., 2017).
NCBI BLASTn analysis revealed that 4 out of the 5 PPIV-1 sequences obtained from SMg had the highest similarity (89.5%-96.2%) to the Chinese PPIV-1 strain S033N (GenBank accession: JX857410.1) ( Table 2), which was first characterized from deceased pigs in Hong Kong in 2013 (Lau et al., 2013). Phylogenetic analysis of BS-1 (MT995732) with other available complete or near-complete genomes revealed clustering with strain S033N (Figure 1c). An additional phylogenetic analysis of partial F and L sequences from BS-1, along with the 3 available PPIV-1 strains from Hungary (Dénes et al., 2020), also showed the closest similarity to strain S033N ( Figure S1). This could indicate that strains genetically related to S033N are established within Central Europe. In comparison, sample NS-2 from a German farm showed the highest similarity (97.0%) to another Chinese PPIV-1 strain, Gd2018 (GenBank accession: MK395271.1), which was collected in 2018.
Previously detected PPIV-1 sequences were obtained from OF, NS and lung samples, with the upper respiratory tract being suggested to be the most suitable sampling site for detection (Agüero et al., 2020;Lau et al., 2013;Park et al., 2019 This is a trade-off between an efficient method to monitor circulating pathogens on a herd level, however, at the same time, pathogens are unable to be linked to an individual animal within a herd.
In conclusion, to the best of our knowledge, we report the first detection of PPIV-1 in Germany and the Netherlands, as well as the first near-complete genome in Europe. Moreover, this is the first detection of PPIV-1 using a targeted and untargeted NGS approach directly from the sample. As the PPIV-1 sequences from Hungary, Germany and the Netherlands were closely related to strains previously found in China, it suggests there may have been PPIV-1 transmission between Europe and China. Furthermore, as PPIV-1 was detected in pigs from 11 different farms (5 using SMg and 6 using RT-qPCR), it could confirm its circulation in Central Europe.
Additional research is required to determine the extent of dissemination of PPIV-1 in Europe, to assess its relevance in the porcine respiratory disease complex and its ability to cross host species barriers.

ACK N OWLED G EM ENTS
We thank Erwin C. RaaNGS for expert technical support and the farmers for their participation in the study.

CO N FLI C T O F I NTE R E S T
John W. A. Rossen is employed by IDbyDNA. Silke Peter consults for IDbyDNA. This did not influence the interpretation of reviewed data and conclusions drawn nor on the drafting of the manuscript, and no support was obtained from them. All other authors declare no conflict of interest.

E TH I C A L A PPROVA L
The animal samples used for this study were collected within the

DATA AVA I L A B I L I T Y S TAT E M E N T
The near-complete PPIV-1 sequence has been deposited under the