TYPE Original Research PUBLISHED 09 February 2023 DOI 10.3389/fmicb.2023.1109881 Infection dynamics, transmission, OPEN ACCESS and evolution after an outbreak of porcine reproductive and EDITED BY Richard John Philip Brown, Paul Ehrlich Institute, Germany REVIEWED BY respiratory syndrome virus Andre Gömer, Ruhr University Bochum, Germany Hepzibar Clilverd *, Gerard Martín-Valls , Yanli Li , Marga Martín , Patrick McClure, Martí Cortey † and Enric Mateu † University of Nottingham, United Kingdom Department of Animal Health and Anatomy, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, Spain *CORRESPONDENCE Hepzibar Clilverd hepzibar.clilverd@uab.cat The present study was aimed at describing the infection dynamics, transmission, and These authors have contributed equally to this † evolution of porcine reproductive and respiratory syndrome virus (PRRSV) after an work and share last authorship outbreak in a 300-sow farrow-to-wean farm that was implementing a vaccination SPECIALTY SECTION This article was submitted to program. Three subsequent batches of piglets (9–11 litters/batch) were followed 1.5 Virology, (Batch 1), 8 (Batch 2), and 12 months after (Batch 3) from birth to 9 weeks of age. The a section of the journal RT-qPCR analysis showed that shortly after the outbreak (Batch 1), one third of sows Frontiers in Microbiology were delivering infected piglets and the cumulative incidence reached 80% by 9 weeks RECEIVED 28November 2022 of age. In contrast, in Batch 2, only 10% animals in total got infected in the same period. ACCEPTED 13 January 2023 PUBLISHED 09 February 2023 In Batch 3, 60% litters had born-infected animals and cumulative incidence rose to 78%. Higher viral genetic diversity was observed in Batch 1, with 4 viral clades circulating, of CITATION Clilverd H, Martín-Valls G, Li Y, Martín M, which 3 could be traced to vertical transmission events, suggesting the existence of Cortey M and Mateu E (2023) Infection founder viral variants. In Batch 3 though only one variant was found, distinguishable dynamics, transmission, and evolution after an from those circulating previously, suggesting that a selection process had occurred. outbreak of porcine reproductive and respiratory syndrome virus. ELISA antibodies at 2 weeks of age were significantly higher in Batch 1 and 3 compared Front. Microbiol. 14:1109881. to Batch 2, while low levels of neutralizing antibodies were detected in either piglets doi: 10.3389/fmicb.2023.1109881 or sows in all batches. In addition, some sows present in Batch 1 and 3 delivered COPYRIGHT infected piglets twice, and the offspring were devoid of neutralizing antibodies at © 2023 Clilverd, Martín-Valls, Li, Martín, Cortey 2 weeks of age. These results suggest that a high viral diversity was featured at the initial and Mateu. This is an open-access article distributed under the terms of the Creative outbreak followed by a phase of limited circulation, but subsequently an escape variant Commons Attribution License (CC BY). The emerged in the population causing a rebound of vertical transmission. The presence of use, distribution or reproduction in other unresponsive sows that had vertical transmission events could have contributed to the forums is permitted, provided the original author(s) and the copyright owner(s) are transmission. Moreover, the records of contacts between animals and the phylogenetic credited and that the original publication in this analyses allowed to trace back 87 and 47% of the transmission chains in Batch 1 and journal is cited, in accordance with accepted 3, respectively. Most animals transmitted the infection to 1–3 pen-mates, but super- academic practice. No use, distribution or reproduction is permitted which does not spreaders were also identified. One animal that was born-viremic and persisted as comply with these terms. viremic for the whole study period did not contribute to transmission. KEYWORDS porcine reproductive and respiratory syndrome virus, genetic diversity, evolution, founder effect, super-spreader, neutralizing antibodies 1. Introduction Porcine reproductive and respiratory syndrome virus (PRRSV) is one of the major pathogens of swine. After its emergence in the 1980 decade, the virus disseminated worldwide and, nowadays, this infection has become endemic in most pig-producing countries. The impact of the infection is variable depending on the virulence of the strain and on the presence of other concomitant agents Frontiers in Microbiology 01 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 [reviewed in Saade et al. (2020) and Zhao et al. (2021)] but, in general, has been related to the immunobiological properties of the strains such the costs associated with the disease are significant for the affected farms as the induced pattern of cytokines (Darwich et al., 2011). Besides this, (Nieuwenhuis et al., 2012; Holtkamp et al., 2013; Renken et al., 2021). recombination events have been linked to the emergence of more PRRSV is an enveloped, positive single-stranded RNA virus that virulent strains (Shi et al., 2013; Chen et al., 2018; Cui et al., 2022). belongs to the genus Betaarterivirus, family Arteriviridae, order Moreover, increased genetic diversity and novel glycosylation sites Nidovirales (Brinton et al., 2021). The 15 kb genome encodes for at least within neutralizing epitopes was observed when vaccination was 10 open reading frames (ORFs). ORF1a and ORF1b encode for the introduced in a herd (Kwon et al., 2019). non-structural proteins (nsp), including the RNA-dependent RNA Although the high number of papers dealing with the genetic polymerase (nsp9). ORF2-4 encode for the minor envelope proteins (E, diversity of the virus, very few have examined how this diversity is GP2, GP3, and GP4), ORF5 for the major envelope glycoprotein (GP5) generated in the affected farms. It is well known that once the farm is and ORF5a protein, ORF6 for the membrane protein (M), and ORF7 for infected, PRRSV most likely will become endemic [reviewed by Pileri the viral nucleocapsid protein (N) (Snijder and Meulenberg, 1998). and Mateu (2016)]. However, in areas of high density of pig farms, At present, two species of PRRSV are recognized, that had been lateral introductions of new strains may happen every few months classified as genotypes until the last taxonomy modification in 2021. because of the widespread presence of the virus and the multiple sources Betaarterivirus suid 1 or PRRSV-1 (formerly called the European of infection. Tousignant et al. (2015) showed that in the United States genotype) was first identified in the Netherlands in early 1990s about one third of the farms suffered a new infection every year. Thus, (Wensvoort et al., 1991), and Betaarterivirus suid 2 or PRRSV-2 one of the most common situations is repeated introductions of different (formerly designated as the North American genotype) was identified PRRSV strains in farms even though sows are vaccinated. This scenario in the United States in 1992 (Benfield et al., 1992). is ideal for the selection of new variants or for the occurrence of PRRSV infection causes reproductive failure in pregnant sows and recombination events. Within the farm, viral selection and diversification respiratory problems, poor growth rate, and mortality in piglets. After could happen mainly: (1) in the sows, the ones which maintain the the contact with the virus, susceptible animals develop viremia, that can infection cycle in the farm through vertical transmission to the offspring, last from days to months depending on the age; the younger the longer. (2) in the piglets, which are infected as soon as they lose the maternally After the cessation of the viremia, infection persists in lymph nodes for derived antibodies, or (3) in both age groups. This is an almost several weeks but finally the infection is cleared. Recovered animals unexplored area. develop strong immunity against the homologous challenge for some The present study explores the genetic diversification of PRRSV over months. In contrast, immunity against the heterologous challenge is time from the ending of a reproductive disease outbreak until 11 months only partial and infection can take place, although the course is often later. We examined the generation of founder viral variants, tracked less severe [revised by Mateu and Diaz (2008)]. transmission chains, assessed the persistence of the variants in the farm, The control of PRRSV infection can be achieved by a combination and evaluated the role of the animals born viremic in the evolution of of monitoring, herd management measures, biosecurity programs, and the virus. The characteristics of different viral variants with regards to vaccination. Commercial vaccines against PRRSV are made either of replication kinetics and the susceptibility to neutralization were attenuated live virus, with or without an adjuvant, or inactivated virus also determined. (always with adjuvant). Attenuated vaccines are preferred for the initial immunization of sows while inactivated vaccines are mostly used as recall antigens. In any case, vaccinated animals can be infected since 2. Materials and methods protection is only partial. In sows, the most common vaccination programs start before the first insemination with the aim of protecting 2.1. Case farm and timeline them before the first gestation. Afterwards, sows need to be revaccinated every 3–4 months. The examined farm housed 300 breeding sows and raised their Very soon after the discovery of the virus it was evident that PRRSV offspring until 9 weeks of age, when they were sent to a fattening unit diversified with a high evolutionary rate. Thus, Forsberg et al. (2001) until market-weight age. The farm had been positive to PRRSV for at proposed that, using the ORF3 to set up a molecular clock, the least 4 years prior to the start of the study. The farm purchased substitution rate was in the range of 4.8 ± 6.9 × 10−3 per site per year, a replacement gilts (young sows that have not farrowed or given birth to value higher than the estimated one for other nidoviruses, for example a litter yet) from a PRRSV-negative herd. Gilts were vaccinated with a coronaviruses (Duchene et al., 2020; Worobey et al., 2020; Ghafari et al., ® modified live PRRSV vaccine (Porcilis PRRS, MSD, Spain) at least 2022; Tay et al., 2022). This, together with the increasing evidence twice before the first gestation and then, together with all sows, received obtained by sequencing, led to the idea of the ever-expanding diversity recall doses at least three times per year (blanket vaccination). With this of PRRSV (Murtaugh et al., 2010; Shi et al., 2010). In addition, the vaccination program the farm had been stabilized and no clinical signs existence of recombination events further boosts the diversification of of PRRS were observed for 2 years until a reproductive outbreak the virus (Brar et al., 2014; Martín-Valls et al., 2014). occurred. The disease was characterized by abortions, stillbirths, and There are multiple consequences of this genetic diversity. On the one weak-born animals. PRRS diagnosis was performed by RT-qPCR, but hand, it is known that genetic diversity has an impact on protection. the viral isolate was different from the one circulating before, indicating From early studies it was evident that heterologous protection was only a lateral introduction of the virus. The timeline and design of the study partial (Mengeling et al., 2003; Labarque et al., 2004). For example, most is shown in Figure 1. strains, induce neutralizing antibodies (Nab) that only significantly A follow-up of the disease and the viral circulation started 6 weeks neutralize the strain that induced them (Martínez-Lobo et al., 2011). On after the onset of the outbreak. The first sampling was conducted one the other hand, genetic diversity also affects the recognition of T-cell and a half months after the onset of the outbreak (Batch 1, 72 piglets epitopes (Vashisht et al., 2008; Díaz et al., 2009; Parida et al., 2012), and from 11 litters), the second one 8 months after the onset of the outbreak Frontiers in Microbiology 02 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 FIGURE 1 Timeline and design of the present study. The figure shows a summary of the follow-up performed in the farm after the onset of a PRRSV-1 reproductive outbreak. Three batches of piglets were followed 1.5 (Batch 1), 8 (Batch 2), and 12 months after the outbreak (Batch 3) from birth to 9 weeks of age. Samples were used to detect PRRSV by RT-qPCR, sequence, and assess the animals’ serological status against PRRSV. (Batch 2, 74 piglets from 9 litters), and the third one 1 year after the onset An internal positive control (IPC) was included in each reaction, in of the outbreak (Batch 3, 70 animals from 10 litters). At least 6 piglets addition to positive and negative controls for the RNA extraction and were selected based on the probability sampling principle within each the PCR (samples with a known viral amount and DEPC-water). litter. In each batch, sows of different parity were included (from parity Samples with Ct values ≤37 were considered positive, and between 37.1 one to eleven overall). The sampling consisted of the collection of and 39.9 were considered inconclusive. With this data, the incidence for umbilical cords (UC) at birth and blood samples at 2, 4, 6, and 9 weeks each observation period was calculated as the proportion of infected of age. All animals included in the follow-up were ear tagged to identify animals over the number of susceptible animals in each time period. them individually. Virus was isolated in porcine alveolar macrophages (PAM) for the subsequent whole genome sequencing. PAM were isolated by bronchoalveolar lavage as reported previously (Mayer and Lam, 1984). 2.2. Detection of porcine reproductive and To avoid biasing the results, only a single passage was carried out. respiratory syndrome virus by RT-qPCR and According to the previous report (Cortey et al., 2018), discrepancies viral isolation between sample-to-sequence and PAM isolate-to-sequence were 1–3 per 10−4 nucleotides with this method. For UC, all positive samples with Ct Umbilical cords (approximately 4 cm) were collected soon after values ≤32.0 were selected for further isolation in PAM. For sera, birth and submerged in a tube containing sterile PBS with antibiotics. isolation was attempted from all animals infected at the farrowing units Scissors and forceps used for sample collection were decontaminated and from their pen-mates that were RT-qPCR positive during the study. after each use by submerging them in 2% sodium hypochlorite and were Cell culture supernatants of the successfully isolated samples were then rinsed with distilled water. The collected UC were minced with collected, aliquoted, and stored at −80°C. sterile scissors and blades, and then the suspension was vortexed and centrifuged at 12,000 × g. The supernatants were aliquoted and stored at −80°C until used. Blood was collected by cava or jugular vein puncture 2.3. Sequencing depending on the age of the animal. Serum was separated by centrifugation after clotting and aliquoted to store at −80°C. For Batch 1 and 2, ORF5, that encodes for the major envelope ® Viral RNA was extracted using the NucleoSpin RNA virus kit protein of the virus and has been traditionally used for phylogenetic (Macherey-Nagel, Germany) according to the manufacturer’s directions, analyses of PRRSV, was amplified and Sanger sequenced for all positive with an elution volume of 50 μL. The presence of PRRSV RNA in the samples with a Ct value ≤32.0 using a previously reported protocol sample was assessed by a commercial RT-qPCR kit aimed to detect both (Mateu et al., 2003). For Batch 3, at least 50% of the samples with the PRRSV-1 and PRRSV-2 (LSI VetMAX™ PRRSV EU/NA kit, Thermo required Ct were sequenced for ORF5 (randomly selected). For Batch 1, Fisher Scientific, United States) according to the instructions provided. viral non-structural protein 2 (nsp2) and nsp9 segments, a highly Frontiers in Microbiology 03 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 variable and a highly conserved segment of the genome, respectively, 2.5. Transmission chains were also partially sequenced using tailor-made oligonucleotide primers (Supplementary Table S1). Since animals were ear-tagged and all movements of animals Whole-genome-based analyses may provide a fuller picture of between pens were recorded, the chain of infection was tracked through PRRSV genetic diversity and evolution. Therefore, viral RNA was the phylogenetic sequences, pens where animals were located, and the extracted from the isolates using the TRIzol™ reagent (Thermo Fisher possible direct contacts animals had before the infection. Traceable Scientific, United States) following the manufacturer’s instructions, with horizontal transmission was considered to occur when the variant in a an elution volume of 20 μL. The viral RNA was used for next generation newly emerged case was the same as the one presented in a direct- sequencing (NGS) using Illumina Miseq without performing any contact pig at the previous observation timepoint. When more than one previous amplification. Briefly, the protocol developed in five steps. First, animal could be considered the source of infection (for example two pigs the genomic library was constructed using a commercial protocol and with the same viral variant had direct contacts), the new case was reagents (Protocol for use with Purified mRNA or rRNA Depleted RNA considered traceable, but the source animal was non-identifiable. With ® and NEBNext Ultra™ II RNA Library Prep Kit for Illumina , New ® this information, maps of transmission were built, and the incidence of England Biolabs, United States). After the NGS run, sequences of low each variant could be estimated approximately. quality were trimmed (QC < 20) using Trimmomatic© (Bolger et al., 2014; RRID:SCR_011848). Then, reads were mapped against a reference sequence (Burrows-Wheeler Aligner applying the BWA-MEM algorithm 2.6. Intrahost diversity for long reads; Li and Durbin, 2010; RRID:SCR_010910). The reference sequence was produced from the earliest available isolate obtained The viral quasi-species inferred from the NGS runs were used to during the outbreak by de novo assembly using SPAdes© (Bankevich analyze the distribution and evolution of the intrahost diversity. Among et al., 2012; RRID:SCR_000131). In the fourth step, variant calling was the individuals sampled, two consecutive quasi-species were available performed with SnpSift© (Cingolani et al., 2012; RRID:SCR_015624) to for five of them. For every viral quasi-species, the nucleotide frequencies determine the frequency of each nucleotide at each position of the per position obtained in the fourth step of the NGS procedure described reference genome. Finally, the viral quasi-species was constructed in in section 2.3 were summarized. The analyses focused on positions with fasta format, and the consensus sequence was obtained using high diversity, arbitrarily defined as those where the frequencies of the Consensus© software, available at www.hiv.lanl.gov/content/sequence/ mutations present were larger than 10%. Finally, the positions identified CONSENSUS/consensus.html. The consensus sequences for the whole were located within the genome. genomes and the ORF5, the nsp2, and the nsp9 sequences were submitted to GenBank with the Accession Numbers OP688189 - OP688223, OP688224 - OP688357, OP822784 - OP822837, and 2.7. Viral replication kinetics OP822838 - OP822859, respectively. The NGS raw sequence reads were also submitted to GenBank with the BioProject ID PRJNA915491. To assess the replication kinetics of the different viral variants found in the farm, one isolate was used of each clade. PAM were inoculated at multiplicity of infection (MOI) of 0.01 or medium for negative controls. 2.4. Sequence analyses Three replicates per clade and negative controls were included for each harvest time-point. Supernatants were harvested at 24, 48, 72, and 96 h For comparative purposes, a selection of available complete genome post-infection. Viral RNA was extracted from the supernatants, and and ORF5 sequences of PRRSV-1 strains including the five PRRSV then quantified by RT-qPCR as described in 2.2. An internal positive vaccines commercially licensed in Spain were downloaded from control was included in all extractions. Ct values were plotted against the Genbank (Supplementary Table S2). Phylogenetic trees were constructed results of a series of dilutions of the same virus. using MrBayes (Bayesian inference with 1,000,000 iterations; Ronquist et al., 2012; RRID:SCR_012067) and p-distances were calculated using MEGA X (Tamura et al., 2021; RRID:SCR_000667). 2.8. Serological analyses For a detailed phylogenetic analysis of the sequences obtained in this study, phylogenetic trees were built up for the whole genome All serum samples in the study were analyzed by ELISA (IDEXX sequence and each viral nsp and ORF using MEGA X (maximum PRRS X3 Ab Test, IDEXX, United States) to determine the animals’ likelihood method, general time reversible model with 1,000 iterations serological status against PRRSV at the moment of infection and the and pairwise deletion) and MrBayes (Bayesian inference with 1,000,000 seroconversion if infected. The ELISA results were expressed as S/P ratio iterations). P-distances were calculated with MEGA X to determine the values, namely, the ratio between the corrected optical density (OD) of a range of inter-and intra-batch and clade diversity for the whole genome given sample (OD sample-OD negative control) over the corrected OD sequences and the viral segments. The substitution rate was estimated of the positive control included in the kit (OD positive control-OD for the whole genome and each segment using BEAST (three negative control). This S/P ratio allowed semi-quantification and independent Markov chain Monte Carlo Bayesian simulations per comparison of the results obtained with the different sera. An S/P segment, 108 steps each; Suchard et al., 2018; RRID:SCR_010228). The ratio ≥ 0.40 was considered positive for PRRSV antibody. Additionally, a existence of potential recombination in the whole genome sequences selection of sera of animals of Batch 1 and 3 was used to assess the was analyzed using GARD (Pond et al., 2006) and RDP5.0 (Martin et al., capability for neutralizing the predominant viral variant at the end of the 2020). In addition, a comparison of the amino acid composition of the study. This selection considered: (i) samples from born-infected animals predicted protein sequences was performed between batches, clades, having viremias longer than 5 weeks, (ii) samples from infected animals and the vaccine used in the farm. at the farrowing units (before weaning) and their PRRSV-negative Frontiers in Microbiology 04 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 siblings, (iii) samples from PRRSV-negative animals at 2 weeks of age, circulation was very low with only 8/74 animals testing positive overall (iv) samples taken at 4 and 9 weeks of age of animals infected at 4 weeks (10.8%; CI95%: 5.1–20.7%, p < 0.05) corresponding to two UC and one, two, of age, or (v) samples from the offspring of sows that were present in the and three animals at 4, 6, and 9 weeks of age, respectively, with Ct values sampling of both Batch 1 and 3 as representative of animals that >31.5. Mortality for this batch was 21.6% (CI95%: 13.2–33.0%). In Batch 3 potentially could have been in contact with the studied viral variants. (n = 70), positive piglets were found in 6 litters at birth (60.0%; CI95%: 27.4– Viral neutralization test (VNT) was performed using the protocol 86.3%; average Ct 35.6 ± 1.0) and results of UC were inconclusive in two described by Yoon et al. (1994) with minor modifications. The virus used other litters. During the observation period, 55 animals (78.6%; CI95%: for this test was the predominant variant in Batch 3 that had been 66.8–87.1%) were infected. Mortality for this batch was 18.6% (p < 0.05). previously adapted to replication in MARC-145 cells and was fully sequenced following the protocol abovementioned. Only NAb titers ≥2 log2 were considered to be relevant. Blood samples of sows (n = 96), which 3.2. Phylogenetic and evolutionary analyses were collected after the present study (2–3 months after Batch 3 sampling) for routine health monitoring purposes, were also examined in the VNT. The phylogenetic analyses of the sequences obtained in this study together with the PRRSV-1 reference sequences and the five vaccines commercially licensed in Spain indicated that the circulating virus 2.9. Statistical analyses strain in the farm of the present study did not derive from the vaccine used in the farm (Supplementary Figures S1, S2) and had a whole The number of PRRSV vertical transmissions and mortality was genome similarity of 86.5% to the vaccine strain compared between batches using a Chi-square test with Yates correction. (Supplementary Table S3). In addition, the analysis showed that S/P ratio values at 2 and 9 weeks of age and at 2 weeks of age based on variants of the same virus strain were circulating throughout the the parity of the sow were compared between batches. Neutralization follow-up and there was no novel introduction of a different virus. titers were log2 normalized and were compared. All statistical analyses A Bayesian phylogenetic tree inferred from the ORF5 sequences were carried out using GraphPad Prism v9 (RRID:SCR_002798). obtained from born-infected animals and the new infection cases is shown in Figure 3. No quality sequences could be obtained from Batch 2 samples. The phylogenetic analysis of ORF5 showed a clear differentiation 3. Results of Batch 1 sequences from Batch 3 sequences. In Batch 1, four significantly different clades were distinguished and 3 of them included variants 3.1. Viral circulation in the farm detected in born-infected animals (animal 515 in Clade 1, animals 526 and 527 in Clade 2, and animals 518–522 in Clade 3). Similar results were Incidence of PRRSV in the different batches is shown in Figure 2. In seen for nsp2 and nsp9 segments (Supplementary Figures S3, S4). The Batch 1, 11 litters including 72 animals were examined. PRRSV-positive analysis of the whole genome sequences confirmed that Batch 1 and Batch animals were detected in 4/11 litters (36.4%; CI95%: 12.4–68.4%) at birth 3 sequences formed two distinct branches (Supplementary Figure S5), (n = 11; between 1 and 5 positive animals per litter; average Ct value and the four abovementioned clades could also be identified in Batch 1. 30.8 ± 4.0). In 1/11 litter, inconclusive RT-qPCR results (Ct > 37.0) were The phylogenetic trees based on maximum likelihood method from obtained from the UC of the new-born piglets, but infection was confirmed ORF5, nsp2, nsp9, and whole genome sequences are shown in at 2 weeks of age. During the observation period of this batch, 58 animals Supplementary Figures S6–S9, respectively. (80.6%; CI95%: 69.2–88.6%) were infected, and 24 (33.3%; CI95%: 22.9– Using ORF5, the net distance between Batch 1 and 3 was 45.5%) died in the period between birth and 9 weeks of age, 10 of which in 0.0135 ± 0.0041 (Table 1). The estimated annual drift for ORF5 was the first 2 weeks of age (41.7% of the total mortality). In Batch 2, viral 1.108% (Table 2). The calculation of the nucleotide mean distance within Batch 1 (p-distance 0.0078 ± 0.0020) was 3 times higher than in Batch 3 (p-distance 0.0023 ± 0.0009). When the whole genome was used for the calculations of the nucleotide mean p-distance within each batch, similar values were obtained (0.0058 ± 0.0004 and 0.0020 ± 0.0002, for Batch 2 and 3 respectively). The estimated annual drift for the whole genome was 0.638%. Furthermore, no recombinant events among the sequences were detected in the analysis. Amino acid composition differences of the viral proteins between clades of Batch 1 and Batch 3 are shown in Supplementary Table S4. All sequences presented a deletion of 5 amino acids in nsp2 between position 347–351 referred to the prototype PRRSV-1 strain Lelystad (LV; NC043487). There were in total 73 amino acid differences between clades of Batch 1, and 61 differences between Batch 1 and 3. Two amino acid variations in GP4 (amino acid positions 65 and 69) and two in GP5 FIGURE 2 (amino acid position 41 and 46) affected known neutralizing epitopes Evolution of the cumulative incidence of PRRSV-1. The graph shows (Meulenberg et al., 1997; Plagemann, 2004; Vanhee et al., 2011). the cumulative incidence of PRRSV-1 for each observation period (2, 4, Additionally, one of these variations in GP5 (amino acid position 46) 6, and 9 weeks of age) per batch. The incidence was calculated as the proportion of infected animals over the number of susceptible animals was located at a known glycosylation site (Meulenberg et al., 1995; in a given time period. Wissink et al., 2003, 2004). When comparing the predicted amino acid composition of the viral proteins between the virus circulating on the Frontiers in Microbiology 05 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 FIGURE 3 Bayesian phylogenetic tree based on the nucleotide sequences of PRRSV-1 ORF5 (gene size of 606bp) using MrBayes (1,000,000 iterations). ORF5 sequences obtained from born-infected animals with long viremias (in red) and the new infection cases (in black) are included. The orange and blue shaded areas correspond to sequences retrieved from Batch 1 and 3, respectively. Sequences from vertical transmission events are marked with a blue triangle. The shaded sequences with the same color correspond to animals belonging to a transmission chain. Only posterior probability values >70% are shown. farm and the used vaccine strain (Supplementary Table S5), there were (Table 3). Also, the positions were not evenly distributed throughout the in total 589 amino acid differences. Thirteen amino acid changes in GP4 genome and the segment corresponding to nsp1a (540 nucleotides) and 2 in GP5 affected known neutralizing epitopes. accumulated most of those highly variable positions, 55% for Batch 1 animals and one third for Batch 3 animals. Of note, no significant differences were observed between the first and the second timepoint 3.3. Intrahost diversity regarding those highly variable positions in each animal. Intrahost diversity was examined at two different timepoints in 5 animals that tested positive for the virus at least twice (2 in Batch 1 and 3.4. Transmission chains 3 in Batch 3). The results showed that the number of highly variable positions was higher in the quasi-species inferred from the sequences of The phylogenetic analyses together with the location records of animals first batch animals (64) compared to the third batch animals (22.5) at each timepoint led to the identification of possible direct contacts of the Frontiers in Microbiology 06 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 TABLE 1 Similarities (p-distance ± standard error) within and between TABLE 2 Yearly drift and 95% confidence interval per segment of the viral Batch 1 and 3. genome. Segment Batch 1 Batch 3 Batch 1 vs 3 Segment 95%-Low % 95%-High Genome 0.0058 ± 0.0004 0.002 ± 0.0002 0.0131 ± 0.0011 Genome 0.434 0.638 0.867 ORF1a ORF1a  nsp1a 0.0057 ± 0.002 0.0037 ± 0.0015 0.0155 ± 0.0049 nsp1a 0.265 0.465 1.029  nsp1b 0.0071 ± 0.0018 0.0024 ± 0.0011 0.0127 ± 0.0041 nsp1b 0.084 0.283 0.627  nsp2 0.006 ± 0.0009 0.0025 ± 0.0005 0.0189 ± 0.0025 nsp2 0.229 0.484 0.826  nsp2TF 0.0068 ± 0.0009 0.003 ± 0.0006 0.0189 ± 0.0022 nsp2TF 0.045 0.109 0.197  nsp3 0.0045 ± 0.0015 0.0008 ± 0.0005 0.0104 ± 0.0031 nsp3 0.311 0.707 1.231  nsp4 0.0088 ± 0.0024 0.0009 ± 0.0005 0.0158 ± 0.0046 nsp4 0.457 0.966 1.674  nsp5 0.0068 ± 0.0022 0.0021 ± 0.001 0.0141 ± 0.0047 nsp5 0.276 1.115 2.352  nsp6 0.0102 ± 0.0103 0±0 0.0027 ± 0.0027 nsp6 0.000 0.006 1.105  nsp7a 0.0092 ± 0.0028 0.0012 ± 0.0007 0.0116 ± 0.0046 nsp7a 0.122 0.562 1.369  nsp7b 0.0033 ± 0.0018 0.001 ± 0.0007 0.0089 ± 0.0046 nsp7b 0.183 0.694 1.523  nsp8 0.0045 ± 0.0024 0.0032 ± 0.0031 0.0039 ± 0.0036 nsp8 0.056 0.501 1.393 ORF1b ORF1b  nsp9 0.0042 ± 0.0009 0.0015 ± 0.0005 0.0124 ± 0.0024 nsp9 0.001 0.361 0.863  nsp10 0.0043 ± 0.0009 0.0015 ± 0.0006 0.0087 ± 0.0023 nsp10 0.351 0.714 1.166  nsp11 0.0019 ± 0.0008 0.0021 ± 0.0007 0.0118 ± 0.0041 nsp11 0.693 1.391 2.299  nsp12 0.0077 ± 0.0022 0.003 ± 0.0012 0.0232 ± 0.0064 nsp12 0.339 1.452 3.091 ORF2a 0.0048 ± 0.0014 0.0012 ± 0.0005 0.0053 ± 0.0022 ORF2a 0.272 0.671 1.218 ORF2b 0.0033 ± 0.0024 0.0009 ± 0.0008 0.0052 ± 0.0044 ORF2b 0.059 0.475 1.284 ORF3 0.0085 ± 0.0021 0.0049 ± 0.0012 0.0076 ± 0.0025 ORF3 0.265 0.527 0.872 ORF4 0.0068 ± 0.0019 0.0017 ± 0.0007 0.0094 ± 0.0037 ORF4 0.309 0.937 1.748 ORF5a 0.0097 ± 0.0056 0.0014 ± 0.0013 0.028 ± 0.0137 ORF5a 0.983 2.759 5.281 ORF5 0.0078 ± 0.002 0.0023 ± 0.0009 0.0135 ± 0.0041 ORF5 0.712 1.108 1.568 ORF6 0.0091 ± 0.0025 0.0017 ± 0.0009 0.0174 ± 0.0052 ORF6 0.266 0.887 1.728 ORF7 0.0053 ± 0.0023 0.0009 ± 0.0008 0.0065 ± 0.0036 ORF7 0.164 0.517 1.041 3UTR 0.0283 ± 0.0078 0.0151 ± 0.0067 0.0023 ± 0.0018 Calculations based on substitution rate estimations obtained by BEAST (Suchard et al., 2018; RRID:SCR_010228), after three independent MCMC Bayesian simulations per segment, 108 5UTR 0.0056 ± 0.0025 0±0 0.0185 ± 0.0088 steps each, considering the default prior parameters provided by the software. The table shows the nucleotide mean differences within and between the examined batches of PRRSV-1 whole genome consensus sequences and each segment of the viral genome. Nsp, non-structural protein. 3.5. Viral replication kinetics infected individuals (shown in Supplementary Material Animation). For To evaluate if differences in transmission were related to the Batch 1, the transmission chain was possible to be traced back in 51 out of replication capability of the viral variants, replication kinetics were the 58 infection cases (87.9%), while for Batch 3, 16 out of the 34 (47.1%) examined for one isolate of each clade (Figure 4). The different clades sequenced cases. In Batch 1, nine animals were identified as the source of displayed generally a similar replicative fitness in cell culture. The isolate 24 out of 28 horizontal transmission events, and 4/9 of these animals of Clade 2 might have shown a higher early infectivity, which could derived from vertical transmission events. In Batch 3, all identified suggest a higher affinity to the viral receptors, although it did not result transmissions could be traced to 8 animals. In both batches, most of the in a higher predominance among the population. identified virus donors transmitted the infection to 1–3 animals; however, in Batch 1, one animal (number 502) was identified as the source of infection for 8 cases, and in Batch 3, one animal (number 777) was 3.6. Serological analyses identified as the source of infection for 4 cases. It is worth noting that in Batch 1, 4 animals that were born viremic 3.6.1. ELISA antibodies remained so until the end of the observation period. Of these, three When the presence of anti-PRRSV antibodies was assessed by contributed to infect pen-mates but the animal number 515 was not ELISA using the sera collected from piglets at 2 and 9 weeks of age, identified as the source of any transmission. Moreover, this animal significant differences in the S/P ratio values were noticed. Thus, for harbored a viral variant from birth to 6 weeks of age and a different one 2-week-old piglets, the highest S/P ratios were found in animals from at 9 weeks of age, suggesting a new infection with a variant horizontally Batch 1, followed by the animals of Batch 3, and the animals of Batch 2 transmitted by another animal. (Figure 5). At 9 weeks of age, S/P ratios were similar between animals of Frontiers in Microbiology 07 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 TABLE 3 Summary of highly variable positions (positions showing overall mutation frequencies higher than 10%) in the 10 viral quasi-species analyzed, and location of those positions in the PRRSV-1 genome. Batch Batch 1 Batch 3 Animal 518 521 769 770 798 Age at 4 weeks 6 weeks 2 weeks 4 weeks 6 weeks 9 weeks 6 weeks 9 weeks 6 weeks 9 weeks sampling ORF1a  nsp1a 29 33 38 27 8 2 7 2 3 3  nsp1b 1 0 0 2 0 0 0 3 1 0  nsp2 17 2 6 9 2 0 1 17 4 7  nsp3 3 0 3 0 0 0 0 3 0 0  nsp4 2 0 1 1 1 0 0 2 0 2  nsp5 4 0 0 2 0 0 0 3 3 0  nsp6 0 0 0 0 0 0 0 0 0 0  nsp7a 6 1 1 0 1 0 0 0 0 1  nsp7b 1 0 0 1 0 0 0 0 1 0  nsp8 1 0 0 0 0 0 0 2 0 1 ORF1b  nsp9 5 1 3 6 0 0 2 4 0 2  nsp10 3 0 1 3 0 0 0 3 0 0  nsp11 0 0 0 0 1 0 1 0 1 0  nsp12 2 0 5 0 0 0 0 0 1 0 ORF2a 1 1 1 1 1 1 0 9 8 1 ORF2b 0 0 0 0 0 0 0 0 0 0 ORF3 1 1 0 1 2 0 1 3 2 1 ORF4 0 0 1 0 0 0 0 2 0 1 ORF5 5 0 3 4 0 0 0 0 1 2 ORF5a 3 0 2 1 0 0 0 0 0 1 ORF6 6 0 0 2 1 0 0 0 0 0 ORF7 0 0 0 2 0 0 0 0 2 1 Number of highly 90 39 65 62 17 3 12 53 27 23 variable positions The table shows the results for two animals of Batch 1 and three animals of Batch 3 that tested positive in two timepoints. Batch 1 and 3, but almost all pigs in Batch 2 were seronegative, indicating that PRRSV circulation was limited in the nurseries at that time. When the 2-week-old piglets in Batch 2 were examined based on the parity of the sow, the statistical analysis indicated that the offspring of sows ≤6 parity had significantly lower antibody levels (Supplementary Figure S4). 3.6.2. Neutralizing antibodies The levels of NAb were examined in different groups of animals. Regarding animals that were born-infected (n = 7), most of them remained negative all over the observation period and only one was detected as positive at week 6 of age (2 log2). The second examined group were those animals that were infected in the farrowing units (n = 16) compared to their siblings (n = 21). Most of the animals in both FIGURE 4 groups were negative, indicating NAb were not related to their infection Virus replication kinetics of the different viral clades identified in the farm. status. Similarly, the examined uninfected animals at 2 weeks of age did The graph shows the kinetic replication curves of variants (n=4) from each clade of Batch 1 and the variant (n=1) in Batch 3 based on Ct values of the not have relevant NAb titers. In addition, average titers at 2 weeks of age supernatants harvested at 24-, 48-, 72-, and 96-h post-infection (hpi) of were not significantly different when comparing Batch 1 and 3. Next, porcine alveolar macrophages (PAM) at multiplicity of infection of 0.01. we examined if animals infected at 4 weeks of age had developed NAb Frontiers in Microbiology 08 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 A B FIGURE 5 S/P ratios of the piglets as determined by ELISA. Each dot represents an examined individual. S/P ratio values ≥0.4 are considered positive (dotted line). The graphs show the S/P ratio values of the piglets at 2 (A) and 9 (B) weeks of age in each of the examined batches. ns=not significant. TABLE 4 Parity, PRRSV-1 vertical transmission, and neutralizing antibodies titers of the offspring (2 weeks of age) of six sows that were present in both Batch 1 and Batch 3. Parities Vertical transmission N° piglets examined VNT titers Sow number Batch 1 vs. Batch 3 Yes/No, batch Batch 1 vs. Batch 3 Batch 1 Batch 3 512 9–11 Yes, Batch 3 3–5 Neg. Neg. 765 4–6 Yes, Batch 1 & 3 4–6 Neg. Neg. 783 4–6 Yes, Batch 1 4–5 Neg. to 5 log2 Neg. 552 9–11 No 4–6 Neg. to 3 log2 Neg. to 2 log2 635 7–9 No 3–6 2–3 log2 Neg. to 5 log2 893 1–3 No 5–4 Neg. to 4 log2 Neg. Titers are shown as log2 values. Neg., negative. 5 weeks later (n = 17). Of the 17 examined animals, only 3 (17.6%) had Finally, we examined the presence of NAb in sera collected from sows developed NAb against the circulating strain of PRRSV (one in Batch 1 for health monitoring purposes over one year after the original outbreak. At and two in Batch 3). Figure 5 summarizes the results of the that moment, most of the sows had low VNT titers or were negative VNT analyses. (Figure 6) but no differences could be determined regarding parity. In addition, we assessed NAb in the offspring (2 weeks of age) of six sows sampled in both Batch 1 and Batch 3 to ascertain whether maternal-to-fetal transfer of NAb increased with time after potential 4. Discussion contact with different viral variants circulating in the farm. Of those six sows, one gave birth to infected piglets in Batch 1, a second gave birth The introduction of PRRSV in a naïve farm usually results in a to infected piglets in Batch 3, and a third gave birth to infected piglets major reproductive outbreak. After a variable period, the reproductive in Batch 1 and 3 (Table 4). When the offspring of the sows were performance improves, and the farm enters in an endemic state in which examined in Batch 1, positive VNT titers were found in four out of six the disease is mostly seen in nurseries. It is commonly assumed that the litters (11/21 animals within the positive litters, 52.4%) with titers vertical transmission is the key element for the maintenance of the ranging from 2 log2 to 5 log2. When the offspring of the same sows were infection in the farm since it results in the birth of infected animals. examined in Batch 3, only 2 litters had positive animals (8/12, 66.7% These piglets can be viremic for a long period and are thought to within the positive litters) with titers ranging from 2 log2 to 5 log2. It is be efficient transmitters of the disease to their pen-mates. Most of the worth noting that the offspring of 2 of the 3 sows that delivered infected control strategies for the successful elimination of PRRSV from the farm piglets at birth tested negative in the VNT in both Batch 1 and 3. rely on depopulation or on strategies like the load, immunization, and Moreover, the offspring of one sow of parities 9 to 11 (Batch 1 and Batch closure of the farm. In other words, it is assumed that after immunizing 3, respectively) always tested negative. all the animals present in the farm and closing the farm to the entrance Frontiers in Microbiology 09 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 A B C D FIGURE 6 Virus neutralization titers in the different groups. Each dot represents an examined individual. Titers are shown as log2 values and only NAb titers ≥2 log2 were considered to be relevant. (A) The graph compares the VNT titers of the PCR positive (PCR+ve) animals at 2 weeks of age and their PCR negative (PCR-ve) siblings. (B) The graph shows the VNT titers of the PCR negative (PCR-ve) animals at 2 weeks of age in Batch 1 and 3. (C) The graph shows the VNT titers of 2-week-old piglets of Batch 1 and 3. (D) The graph shows the VNT titers according to the parity of the 96 sows sampled one year after the onset of the outbreak. ns=not significant. of new susceptible animals, the circulation of the infection will cease. We started the study approximately 6 weeks after the initial detection of The duration of the closing period is a key element in those strategies. the outbreak. At that moment about one third of the examined litters had However, in many farms, depopulation or closure strategies are not infected animals at birth, indicating a high proportion of susceptible sows. economically feasible and other approaches are needed. The phylogenetic analysis at that point showed a high viral diversity with 4 In those other cases, monitoring, vaccination, and biosecurity recognizable different clades of the virus. Three of these clades were measures are essential for the control. Monitoring usually focuses on the identified in new-borns, suggesting the existence of a founder effect. These detection of infected animals and the hotspots of viral circulation founder effects were described initially in HIV and hepatitis C virus and can (replacement gilts, sows, and nurseries). Additionally, it uses sequencing represent population or selection bottlenecks (Weiner et al., 1993; Gerotto as a tool to determine if the viral circulation in the farm is caused by the et al., 2006; Russell et al., 2011; Naidoo et al., 2017). In other words, the previously resident strain or by a newly introduced one. Sometimes the transmission of the virus from one individual to the other through mucosal same strain reemerges after a period in which viral circulation seemed surfaces or through the placenta may result in a quantitative selection of to be contained. Our knowledge of the microepidemiology of PRRSV viral particles (only few can pass through) or in a qualitative selection (only within the farm over time is still limited. some viral particles with specific characteristics can be transmitted through The present study focused on the examination of the diversification the mucosal surfaces). In our case, we cannot know the nature of these of the virus after an outbreak in a vaccinated farm, relating the viral bottlenecks but the fact that the sows had been multiple vaccinated and evolution with the variations in the incidence of the infection and the might have been infected previously suggest that both types of bottlenecks development of an immune response to the virus. For that purpose, could have acted. In a previous work (Cortey et al., 2018) we have shown we followed a farm for one year after a reproductive outbreak. that selection bottlenecks for PRRSV can exist in vaccinated animals. Frontiers in Microbiology 10 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 After the initial outbreak and the high spread of the infection in the many sows did never encounter the wild-type virus and, in consequence, farrowing units and nurseries in the first batch, in Batch 2 (around did not produce antibodies specific for the farm strain. It is tempting to 6 months later), the viral circulation was limited. The few animals in this think that they are unresponsive sows. This hypothesis would batch that tested positive in the RT-qPCR produced very high Ct values, be supported by the observation that the offspring of sows that indicating low viral loads or even detection of environmental virus for transmitted the infection at birth did not have Nabs, not even in the case example in the case of UC collection. The key question is why the viral of an old sow delivering infected piglets twice. These breeders are clear circulation almost stopped. The most intuitive answer would be that the candidates to maintain infection on the farm. Elucidating the role of circulation ceased because the population was immune. However, when those poor responder sows and determining if the lack of response the animals of that batch were serologically examined, the antibody levels occurred from the initial immunization or was the result of receiving were low in 2-week-old piglets, suggesting that the sows did not have a multiple vaccine doses would be important to understand how the virus strong humoral immunity against the virus. Whether this was because of may persist in vaccinated farms. However, the fact that the reproductive a rapid decline of herd immunity after the outbreak or because of the poor disease was under control after the initial outbreak, and no changes in induction of antibodies by that particular virus isolate, or both, is difficult the farm management or the vaccination schedule were performed, to say. In any case, the absence of antibodies against PRRSV in most of the suggests that some immunity existed in sows. If NAb specific for the 9-week-old animals of Batch 2 is a strong indication that the virus was not farm strain were low, the most logical explanation would be that cell- circulating indeed. Moreover, as we will discuss later, the fact that both mediated immunity had a role in this control. However, the fact that in piglets and sows had low levels of NAb against the circulating virus suggests Batch 3 vertical transmission was high opens the question of the that humoral immunity might not be a complete indicator for duration of the efficacious immunity after infection with some strains. understanding how immunity affects viral persistence in the herd. Taken together, in our opinion, the most likely scenario was as Although correlates of protection against PRRSV have not been fully follows: the introduction of a new strain in the farm found a population elucidated, Lopez et al. (2007) showed that passive transfer of neutralizing that had limited immunity against that strain despite vaccination. At that antibodies results in protection in a homologous challenge model when moment, the infection spread rapidly, affected pregnant sows, and they titers of 1:8–1:16 are reached. However, there is evidence that cell mediated transmitted the infection vertically to the offspring. This vertical immunity also plays a role in protection against PRRSV-1 infection in the transmission involved several bottlenecks (transmission through homologous and heterologous models (Díaz et al., 2012). placenta, transmission to vaccinated sows) that generated, from the Then, we examined Batch 3 four months after Batch 2. Surprisingly, the replication of founder variants, a higher diversity circulating in the infection not only did not disappear but re-emerged, and at the end of the piglets, a fact that suggested that the bottleneck selection had a nursery period most animals had been infected. How did this happen? The component of probability derived from different phenomena (population phylogenetic analysis of Batch 3 isolates clearly showed that a new variant bottlenecks, genetic bottlenecks, etc.). In a subsequent phase, the of the resident virus gained predominance. However, the variant found at population became immune and limited the transmission to a few this moment was much less diverse than any variant found in Batch 1, individuals. However, since the strain infecting the animals in the farm indicating a strong selection process that happened very close to the apparently was a poor inducer of immunity (low levels of NAb and rapid timepoint when Batch 3 was examined. This could reflect the selection of decay of these), susceptible subpopulations rapidly appeared in which the a variant that was able to escape the immune response and that was virus kept circulating until a fitter variant was selected and spread again transmitted efficiently in the vaccinated population that had been infected in the population. At this point, all previous diversity would disappear or in contact with the initial one. One evident explanation could have been and would be replaced by the new variant. An interesting question raises the selection of a variant with enhanced replication or capacity of here, is it possible to identify beforehand what animals contribute to the transmission. Our replication kinetics experiments however did not maintenance of the virus in that phase of limited circulation? This is an support the hypothesis of an enhanced replication capability, although area that needs to be explored in further studies. we cannot discard higher potential for transmission, not caused by an Interestingly, when highly variable positions were examined within the increased yield of replication itself, but by a better replication in the nasal viral quasi-species of animals in Batch 1 and 3, the number of highly mucosa macrophages (Frydas and Nauwynck, 2016). variable positions decreased in Batch 3, suggesting that selection reduced Interestingly, in Batch 3 the proportion of vertical transmission the variability in those highly variable positions. Furthermore, in both events was very high suggesting again that either immunity against the Batch 1 and 3, a high proportion (30 to 55%) of the highly variable new variant was not strong enough in the population or that positions within the quasi-species were in nsp1a despite that nsp1a only non-immune sow subpopulations existed. When we examined the NAb accounts for roughly 3.5% of the viral genome. Nsp1a is involved in the against the variant predominating in Batch 3, it was evident that even downregulation of type 1 interferon (Beura et al., 2010; Chen et al., 2010; sows that had been infected (as demonstrated by the delivery of infected Beura et al., 2012). This activity has been located in the PCPα domain of piglets) developed very low levels of NAb. This is an indication that the the protein. Of note, almost all the highly variable positions in nsp1a were strain present in the farm was probably a poor inducer of NAb. found outside that area, in the zinc-finger motif, suggesting that variability Moreover, the fact that the offspring of old sows that had been vaccinated was not related with the inhibition of the interferon response. Besides, on multiple occasions and had given birth to infected piglets were nsp1a has been shown to contain at least two epitopes inducing negative in the neutralization tests indicates the vaccination did not IFN-gamma responses in CD8 T cells (Mötz et al., 2022). One of the most contribute to enhance the humoral response of the sows against an highly variable positions affected one of them but it is difficult to predict unrelated strain. Furthermore, it is evident that subpopulations of the result of this variability on the effective immunity against the virus. susceptible sows can be found in multiply vaccinated farms with high Moreover, nsp1a is involved in the induction of apoptosis, circulation of wild type virus (Díaz et al., 2020; Fiers et al., 2022). In a pro-inflammatory responses, and the regulation of minus strand templates small farm like the one studied here, after an outbreak and with vertical (Nedialkova et al., 2010; Wang et al., 2016; Park et al., 2021) and thus the transmission rates of 30%, it seems unlikely that solely by chance so variability may respond to other functionalities of the protein. Frontiers in Microbiology 11 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 The present study also provides some relevant information about Ethics statement transmission. The identification of animals, viral variants, and movements between pens allowed to determine, at least partially, the transmission The animal study was reviewed and approved by Ethics chains. These data showed that while most infected animals contributed to Committee in Animal and Human Research of the Universitat 1–3 transmissions, some animals apparently did not transmit the infection Autònoma de Barcelona. Written informed consent was obtained to anyone else, and one (animal 502) was identified as the source of 8 cases from the owners for the participation of their animals in this out of 28 horizontal transmission events (28.6%) behaving like a super- study. spreader. Whether this superspreading capability derived from host characteristics, environmental conditions, or both, cannot be determined. It was recently suggested that superspreading events can be an intrinsic Author contributions characteristic of each viral infection (Chen et al., 2021). It is not known if different viral variants may be more prone to produce superspreading MM, MC, and EM designed the study. Laboratory work was events, but, in our opinion, this would not be the case in our study since performed by HC, YL, GM-V, MC, and EM. EM, MC, GM-V, YL, and other animals infected by the same variant transmitted the infection only HC analyzed the results. EM, MC, and HC prepared the draft of the to one susceptible recipient. manuscript. All authors have read, revised, discussed, and agreed to the Another surprising observation was the fact that one animal that published version of the manuscript. was born-viremic and persisted as so for the whole length of the study, apparently did not contribute to the transmission of the virus. In general, it is thought that animals infected with PRRSV in utero are important Funding spreaders due to the long viremias they develop (Rowland et al., 1999; Benfield et al., 2000; Rowland et al., 2003). However, the evidence This study was funded by the Spanish Ministry of Economy and provided by this study indicates that, at least in some cases, these Competitiveness (project AGL2017-87073-R). HC is supported by a animals may not contribute significantly to the transmission. It would pre-doctoral fellowship Ayudas para la formación de profesorado be interesting to examine the shedding of the virus in oral fluids and universitario (FPU) of the Spanish Ministry of Universities (grant nasal secretions of these born-viremic animals compared to animals number FPU18/04259). MC was supported by a Ramón y Cajal contract infected after birth to determine differences in transmission capabilities. of the Spanish Ministry of Economy and Competitiveness (grant From the practical point of view, it is also worth commenting the number RyC-2015-17154). results of Batch 2. Commonly, PRRSV stability is determined after the examination by RT-qPCR of a number of animals at weaning (usually 30 to 60 animals at wean, in four consecutive batches; Holtkamp, 2011). Acknowledgments In our case, only one sample out of 62 (1.6%) was positive at weaning with a Ct of 31.9. This suggests that in endemic situations sampling to Our grateful thanks to the staff members of the farm, Núria del Val determine stability of a herd should be of at least 60 animals, and and Cooperativa Plana de Vic for their collaboration, and Jordi Soto for probably more. The low prevalence at weaning in that batch was enough his help in the sampling. to allow the reemergence of the virus some weeks later. Conflict of interest 5. Conclusion The authors declare that the research was conducted in the absence The present paper reports an example of how PRRSV may persist in of any commercial or financial relationships that could be construed as a vaccinated farm. Upon introduction of a new strain, founder effects a potential conflict of interest. related to transplacental transmission and high rates of horizontal transmission produce a high diversity of viral variants, but it simultaneously generates an immune or partially immune population, Publisher’s note resulting in a low-level circulation of the virus. The combination of subpopulations that just by chance did not have contact with the virus All claims expressed in this article are solely those of the authors together with a viral strain that was a poor inducer of long-lasting and do not necessarily represent those of their affiliated immunity, kept the virus circulating at a low level. Nevertheless, this was organizations, or those of the publisher, the editors and the enough to allow the selection of a variant that became predominant reviewers. Any product that may be evaluated in this article, or afterwards. The role of super-spreaders and long-term viremic animals claim that may be made by its manufacturer, is not guaranteed or must be re-examined after the results of the present study. endorsed by the publisher. Data availability statement Supplementary material The datasets presented in this study can be found in online The Supplementary material for this article can be found online at: repositories. The names of the repository/repositories and accession https://www.frontiersin.org/articles/10.3389/fmicb.2023.1109881/ number(s) can be found in the article/Supplementary material. full#supplementary-material Frontiers in Microbiology 12 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 References Bankevich, A., Nurk, S., Antipov, D., Gurevich, A. A., Dvorkin, M., Kulikov, A. S., et al. Frydas, I. S., and Nauwynck, H. J. (2016). Replication characteristics of eight virulent (2012). SPAdes: a new genome assembly algorithm and its applications to single-cell and two attenuated genotype 1 and 2 porcine reproductive and respiratory syndrome virus sequencing. J. Comput. Biol. 19, 455–477. doi: 10.1089/CMB.2012.0021 (PRRSV) strains in nasal mucosa explants. Vet. Microbiol. 182, 156–162. doi: 10.1016/J. VETMIC.2015.11.016 Benfield, D. A., Nelson, E., Collins, J. E., Harris, L., Goyal, S. M., Robison, D., et al. (1992). Characterization of swine infertility and respiratory syndrome (SIRS) virus (isolate Gerotto, M., Resti, M., Dal Pero, F., Migliorato, I., Alberti, A., and Bortolotti, F. (2006). ATCC VR-2332). J. Vet. Diagn. Investig. 4, 127–133. doi: 10.1177/104063879200400202 Evolution of hepatitis C virus Quasispecies in children with chronic hepatitis C. Infection 34, 62–65. doi: 10.1007/S15010-006-5037-4 Benfield, D., Nelson, E., Rossow, K., Nelson, J., Steffen, M., and Rowland, R. (2000). Diagnosis of persistent or prolonged porcine reproductive and respiratory syndrome virus Ghafari, M., Du Plessis, L., Raghwani, J., Bhatt, S., Xu, B., Pybus, O. G., et al. (2022). infections. Vet. Res. 31:71. doi: 10.1051/vetres:2000007 Purifying selection determines the short-term time dependency of evolutionary rates in SARS-CoV-2 and pH1N1 influenza. Mol. Biol. Evol. 39:msac009. doi: 10.1093/MOLBEV/ Beura, L. K., Sarkar, S. N., Kwon, B., Subramaniam, S., Jones, C., Pattnaik, A. K., et al. MSAC009 (2010). Porcine reproductive and respiratory syndrome virus nonstructural protein 1beta modulates host innate immune response by antagonizing IRF3 activation. J. Virol. 84, Holtkamp, D. J. (2011). Terminology for classifying swine herds by porcine reproductive 1574–1584. doi: 10.1128/JVI.01326-09 and respiratory syndrome virus status. J. Swine Health Prod. 19, 44–56. doi: 10.1017/ Beura, L. K., Subramaniam, S., Vu, H. L., Kwon, B., Pattnaik, A. K., and Osorio, F. A. S1466252310/00010 (2012). Identification of amino acid residues important for anti-IFN activity of porcine Holtkamp, D. J., Kliebenstein, J. B., Neumann, E. J., Zimmerman, J. J., Rotto, H. F., reproductive and respiratory syndrome virus non-structural protein 1. Virology 433, Yoder, T. K., et al. (2013). Assessment of the economic impact of porcine reproductive and 431–439. doi: 10.1016/j.virol.2012.08.034 respiratory syndrome virus on United States pork producers. J. Swine Health Prod. 21, Bolger, A. M., Lohse, M., and Usadel, B. (2014). Trimmomatic: a flexible trimmer for 72–84. Illumina sequence data. Bioinformatics 30, 2114–2120. doi: 10.1093/BIOINFORMATICS/ Kwon, T., Yoo, S. J., Lee, D. U., Sunwoo, S. Y., Je, S. H., Park, J. W., et al. (2019). Differential BTU170 evolution of antigenic regions of porcine reproductive and respiratory syndrome virus 1 Brar, M. S., Shi, M., Hui, R. K. H., and Leung, F. C. C. (2014). Genomic evolution of before and after vaccine introduction. Virus Res. 260, 12–19. doi: 10.1016/j. porcine reproductive and respiratory syndrome virus (PRRSV) isolates revealed by deep virusres.2018.11.004 sequencing. PLoS One 9:e88807. doi: 10.1371/JOURNAL.PONE.0088807 Labarque, G., Van Reeth, K., Nauwynck, H., Drexler, C., Van Gucht, S., and Pensaert, M. Brinton, M. A., Gulyaeva, A. A., Balasuriya, U. B. R., Dunowska, M., Faaberg, K. S., (2004). Impact of genetic diversity of European-type porcine reproductive and respiratory Goldberg, T., et al. (2021). ICTV virus taxonomy profile: Arteriviridae 2021. J. Gen. Virol. syndrome virus strains on vaccine efficacy. Vaccine 22, 4183–4190. doi: 10.1016/j. 102:001632. doi: 10.1099/jgv.0.001632 vaccine.2004.05.008 Chen, P. Z., Koopmans, M., Fisman, D. N., and Gu, F. X. (2021). Understanding why Li, H., and Durbin, R. (2010). Fast and accurate long-read alignment with burrows– superspreading drives the COVID-19 pandemic but not the H1N1 pandemic. Lancet Infect. wheeler transform. Bioinformatics 26, 589–595. doi: 10.1093/BIOINFORMATICS/BTP698 Dis. 21, 1203–1204. doi: 10.1016/S1473-3099(21)00406-0 Lopez, O. J., Oliveira, M. F., Garcia, E. A., Kwon, B. J., Doster, A., and Osorio, F. A. (2007). Chen, Z., Lawson, S., Sun, Z., Zhou, X., Guan, X., Christopher-Hennings, J., et al. (2010). Protection against porcine reproductive and respiratory syndrome virus (PRRSV) infection Identification of two auto-cleavage products of nonstructural protein 1 (nsp1) in porcine through passive transfer of PRRSV-neutralizing antibodies is dose dependent. Clin. Vaccine reproductive and respiratory syndrome virus infected cells: nsp1 function as interferon Immunol. 14, 269–275. doi: 10.1128/CVI.00304-06 antagonist. Virology 398, 87–97. doi: 10.1016/j.virol.2009.11.033 Martin, D. P., Varsani, A., Roumagnac, P., Botha, G., Maslamoney, S., Schwab, T., et al. (2020). Chen, N., Ye, M., Li, S., Huang, Y., Zhou, R., Yu, X., et al. (2018). Emergence of a novel RDP5: a computer program for analyzing recombination in, and removing signals of highly pathogenic recombinant virus from three lineages of porcine reproductive and recombination from, nucleotide sequence datasets. Virus Evol. 7:87. doi: 10.1093/ve/veaa087 respiratory syndrome virus 2 in China 2017. Transbound. Emerg. Dis. 65, 1775–1785. doi: Martínez-Lobo, F., Díez-Fuertes, F., Simarro, I., Castro, J. M., and Prieto, C. (2011). 10.1111/TBED.12952 Porcine reproductive and respiratory syndrome virus isolates differ in their susceptibility Cingolani, P., Patel, V. M., Coon, M., Nguyen, T., Land, S. J., Ruden, D. M., et al. (2012). to neutralization. Vaccine 29, 6928–6940. doi: 10.1016/j.vaccine.2011.07.076 Using Drosophila melanogaster as a model for genotoxic chemical mutational studies with Martín-Valls, G. E., Kvisgaard, L. K., Tello, M., Darwich, L., Cortey, M., a new program. SnpSift. Front. Genet. 3:35. doi: 10.3389/FGENE.2012.00035 Burgara-Estrella, A. J., et al. (2014). Analysis of ORF5 and full-length genome sequences Cortey, M., Arocena, G., Pileri, E., Martín-Valls, G., and Mateu, E. (2018). Bottlenecks of porcine reproductive and respiratory syndrome virus isolates of genotypes 1 and 2 in the transmission of porcine reproductive and respiratory syndrome virus (PRRSV1) to retrieved worldwide provides evidence that recombination is a common phenomenon and naïve pigs and the quasi-species variation of the virus during infection in vaccinated pigs. may produce mosaic isolates. J. Virol. 88, 3170–3181. doi: 10.1128/JVI.02858-13 Vet. Res. 49, 107–111. doi: 10.1186/S13567-018-0603-1 Mateu, E., and Diaz, I. (2008). The challenge of PRRS immunology. Vet. J. 177, 345–351. Cui, X.-Y., Xia, D.-S., Huang, X.-Y., Tian, X.-X., Wang, T., Yang, Y.-B., et al. (2022). doi: 10.1016/j.tvjl.2007.05.022 Recombinant characteristics, pathogenicity, and viral shedding of a novel PRRSV variant Mateu, E., Martín, M., and Vidal, D. (2003). Genetic diversity and phylogenetic analysis derived from twice inter-lineage recombination. Vet. Microbiol. 271:109476. doi: 10.1016/j. of glycoprotein 5 of European-type porcine reproductive and respiratory virus strains in vetmic.2022.109476 Spain. J. Gen. Virol. 84, 529–534. doi: 10.1099/VIR.0.18478-0/CITE/REFWORKS Darwich, L., Gimeno, M., Sibila, M., Diaz, I., de la Torre, E., Dotti, S., et al. (2011). Mayer, P., and Lam, C. (1984). Porcine alveolar macrophages, isolation, morphological Genetic and immunobiological diversities of porcine reproductive and respiratory and functional characteristics. Zentralbl. Veterinarmed. A 31, 59–71. doi: 10.1111/ syndrome genotype I strains. Vet. Microbiol. 150, 49–62. doi: 10.1016/j.vetmic. j.1439-0442.1984.tb01262.x 2011.01.008 Mengeling, W. L., Lager, K. M., Vorwald, A. C., and Koehler, K. J. (2003). Strain specificity of Díaz, I., Genís-Jorquera, B., Martín-Valls, G. E., and Mateu, E. (2020). Using commercial the immune response of pigs following vaccination with various strains of porcine reproductive ELISAs to assess humoral response in sows repeatedly vaccinated with modified live and respiratory syndrome virus. Vet. Microbiol. 93, 13–24. doi: 10.1016/S0378-1135(02)00427-3 porcine reproductive and respiratory syndrome virus. Vet. Rec. 1 186:123. doi: 10.1136/ vr.105432 Meulenberg, J. J., Petersen-den Besten, A., De Kluyverer, E. P., Moormann, R. J., Schaaper, W. M., and Wenswoort, G. (1995). Characterization of proteins encoded by ORFs Díaz, I., Gimeno, M., Darwich, L., Navarro, N., Kuzemtseva, L., López, S., et al. (2012). 2 to 7 of Lelystad virus. Virology 206, 155–163. doi: 10.1016/s0042-6822(95)80030-1 Characterization of homologous and heterologous adaptive immune responses in porcine reproductive and respiratory syndrome virus infection. Vet. Res. 9 43:30. doi: Meulenberg, J. J., van Nieuwstadt, A. P., van Essen-Zandbergen, A., and Langeveld, J. P. 10.1186/1297-9716-43-30 (1997). Posttranslational processing and identification of a neutralization domain of the GP4 protein encoded by ORF4 of Lelystad virus. J. Virol. 71, 6061–6067. doi: 10.1128/ Díaz, I., Pujols, J., Ganges, L., Gimeno, M., Darwich, L., Domingo, M., et al. (2009). JVI.71.8.6061-6067.1997 In silico prediction and ex vivo evaluation of potential T-cell epitopes in glycoproteins 4 and 5 and nucleocapsid protein of genotype-I (European) of porcine reproductive Mötz, M., Stas, M. R., Hammer, S. E., Duckova, T., Fontaine, F., Kiesler, A., et al. (2022). and respiratory syndrome virus. Vaccine 27, 5603–5611. doi: 10.1016/j. Identification of MHC-I-presented porcine respiratory and reproductive syndrome virus vaccine.2009.07.029 (PRRSV) peptides reveals immunogenic epitopes within several non-structural proteins recognized by CD8+ T cells. Viruses 14:1891. doi: 10.3390/v14091891 Duchene, S., Featherstone, L., Haritopoulou-Sinanidou, M., Rambaut, A., Lemey, P., and Baele, G. (2020). Temporal signal and the phylodynamic threshold of SARS-CoV-2. Virus Murtaugh, M. P., Stadejek, T., Abrahante, J. E., Lam, T. T., and Leung, F. C.-C. (2010). Evol. 6:veaa061. doi: 10.1093/VE/VEAA061 The ever-expanding diversity of porcine reproductive and respiratory syndrome virus. Virus Res. 154, 18–30. doi: 10.1016/j.virusres.2010.08.015 Fiers, J., Tignon, M., Cay, A. B., Simons, X., and Maes, D. (2022). Porcine reproductive and respiratory syndrome virus (PRRSv): a cross-sectional study on ELISA seronegative, Naidoo, V. L., Mann, J. K., Noble, C., Adland, E., Carlson, J. M., Thomas, J., et al. (2017). multivaccinated sows. Viruses 14:1944. doi: 10.3390/v14091944 Mother-to-child HIV transmission bottleneck selects for consensus virus with lower gag- protease-driven replication capacity. J. Virol. 91, 518–535. doi: 10.1128/JVI.00518-17 Forsberg, R., Oleksiewicz, M. B., Krabbe Petersen, A. M., Hein, J., Bøtner, A., and Storgaard, T. (2001). A molecular clock dates the common ancestor of European-type Nedialkova, D. D., Gorbalenya, A. E., and Snijder, E. J. (2010). Arterivirus Nsp1 porcine reproductive and respiratory syndrome virus at more than 10 years before the modulates the accumulation of minus-strand templates to control the relative abundance emergence of disease. Virology 289, 174–179. doi: 10.1006/VIRO.2001.1102 of viral mRNAs. PLoS Pathog. 6:e1000772. doi: 10.1371/journal.ppat.1000772 Frontiers in Microbiology 13 frontiersin.org Clilverd et al. 10.3389/fmicb.2023.1109881 Nieuwenhuis, N., Duinhof, T. F., and Van Nes, A. (2012). Economic analysis of outbreaks Suchard, M. A., Lemey, P., Baele, G., Ayres, D. L., Drummond, A. J., and Rambaut, A. of porcine reproductive and respiratory syndrome virus in nine sow herds. Vet. Rec. (2018). Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10. Virus 170:225. doi: 10.1136/VR.100101 Evol. 4:1. doi: 10.1093/VE/VEY016 Parida, R., Choi, I. S., Peterson, D. A., Pattnaik, A. K., Laegreid, W., Zuckermann, F. A., Tamura, K., Stecher, G., and Kumar, S. (2021). MEGA11: molecular evolutionary et al. (2012). Location of T-cell epitopes in nonstructural proteins 9 and 10 of type-II genetics analysis version 11. Mol. Biol. Evol. 38, 3022–3027. doi: 10.1093/MOLBEV/ porcine reproductive and respiratory syndrome virus. Virus Res. 169, 13–21. doi: 10.1016/J. MSAB120 VIRUSRES.2012.06.024 Tay, J. H., Porter, A. F., Wirth, W., and Duchene, S. (2022). The emergence of SARS- Park, I. B., Choi, Y. C., Lee, K. T., and Chun, T. (2021). Transcriptome analysis of pig CoV-2 variants of concern is driven by acceleration of the substitution rate. Mol. Biol. Evol. macrophages expressing porcine reproductive and respiratory syndrome virus non-structural 39:2. doi: 10.1093/MOLBEV/MSAC013 protein 1. Vet. Immunol. Immunopathol. 231:110147. doi: 10.1016/j.vetimm.2020.110147 Tousignant, S. J. P., Perez, A. M., Lowe, J. F., Yeske, P. E., and Morrison, R. B. (2015). Pileri, E., and Mateu, E. (2016). Review on the transmission porcine reproductive and Temporal and spatial dynamics of porcine reproductive and respiratory syndrome virus respiratory syndrome virus between pigs and farms and impact on vaccination. Vet. Res. infection in the United States. Am. J. Vet. Res. 76, 70–76. doi: 10.2460/AJVR.76.1.70 47, 108–113. doi: 10.1186/S13567-016-0391-4 Vanhee, M., Van Breedam, W., Costers, S., Geldhof, M., Noppe, Y., and Nauwynck, H. Plagemann, P. G. (2004). GP5 ectodomain epitope of porcine reproductive and (2011). Characterization of antigenic regions in the porcine reproductive and respiratory respiratory syndrome virus, strain Lelystad virus. Virus Res. 102, 225–230. doi: 10.1016/j. syndrome virus by the use of peptide-specific serum antibodies. Vaccine 29, 4794–4804. virusres.2004.01.031 doi: 10.1016/j.vaccine.2011.04.071 Pond, S. L. K., Posada, D., Gravenor, M. B., Woelk, C. H., and Frost, S. D. W. (2006). Vashisht, K., Goldberg, T. L., Husmann, R. J., Schnitzlein, W., and Zuckermann, F. A. Automated phylogenetic detection of recombination using a genetic algorithm. Mol. Biol. (2008). Identification of immunodominant T-cell epitopes present in glycoprotein 5 of the Evol. 23, 1891–1901. doi: 10.1093/MOLBEV/MSL051 North American genotype of porcine reproductive and respiratory syndrome virus. Renken, C., Nathues, C., Swam, H., Fiebig, K., Weiss, C., Eddicks, M., et al. (2021). Vaccine 26, 4747–4753. doi: 10.1016/j.vaccine.2008.06.047 Application of an economic calculator to determine the cost of porcine reproductive and Wang, X., Shao, C., Wang, L., Li, Q., Song, H., and Fang, W. (2016). The viral non- respiratory syndrome at farm-level in 21 pig herds in Germany. Porc. Health Manag. 7, structural protein 1 alpha (Nsp1α) inhibits p53 apoptosis activity by increasing murine 1–12. doi: 10.1186/S40813-020-00183-X/FIGURES/3 double minute 2 (mdm2) expression in porcine reproductive and respiratory syndrome Ronquist, F., Teslenko, M., Van Der Mark, P., Ayres, D. L., Darling, A., Höhna, S., et al. virus (PRRSV) early-infected cells. Vet. Microbiol. 184, 73–79. doi: 10.1016/j. (2012). MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a vetmic.2016.01.009 large model space. Syst. Biol. 61, 539–542. doi: 10.1093/SYSBIO/SYS029 Weiner, A. J., Thaler, M. M., Crawford, K., Ching, K., Kansopon, J., Chien, D. Y., et al. Rowland, R. R., Lawson, S., Rossow, K., and Benfield, D. A. (2003). Lymphoid tissue (1993). A unique, predominant hepatitis C virus variant found in an infant born to a tropism of porcine reproductive and respiratory syndrome virus replication during mother with multiple variants. J. Virol. 67, 4365–4368. doi: 10.1128/JVI.67.7.4365-4368.1993 persistent infection of pigs originally exposed to virus in utero. Vet. Microbiol. 96, 219–235. Wensvoort, G., Terpstra, C., Pol, J. M., ter Laak, E. A., Bloemraad, M., de Kluyver, E. P., doi: 10.1016/j.vetmic.2003.07.006 et al. (1991). Mystery swine disease in the Netherlands: the isolation of Lelystad virus. Vet. Rowland, R. R., Steffen, M., Ackerman, T., and Benfield, D. A. (1999). The evolution of Q. 13, 121–130. doi: 10.1080/01652176.1991.9694296 porcine reproductive and respiratory syndrome virus: quasispecies and emergence of a Wissink, E. H. J., Kroese, M. V., Maneschijn-Bonsing, J. G., Meulenberg, J. J. M., van virus subpopulation during infection of pigs with VR-2332. Virology 259, 262–266. doi: Rijn, P. A., Rijsewijk, F. A. M., et al. (2004). Significance of the oligosaccharides of the 10.1006/viro.1999.9789 porcine reproductive and respiratory syndrome virus glycoproteins GP2a and GP5 for Russell, E. S., Kwiek, J. J., Keys, J., Barton, K., Mwapasa, V., Montefiori, D. C., et al. (2011). infectious virus production. J. Gen. Virol. 85, 3715–3723. doi: 10.1099/vir.0.80402-0 The genetic bottleneck in vertical transmission of subtype C HIV-1 is not driven by Wissink, E. H. J., van Wijk, H. A. R., Kroese, M. V., Weiland, E., Meulenberg, J. J. M., selection of especially neutralization-resistant virus from the maternal viral population. J. Rottier, P. J. M., et al. (2003). The major envelope protein, GP5, of a European porcine Virol. 85, 8253–8262. doi: 10.1128/JVI.00197-11 reproductive and respiratory syndrome virus contains a neutralization epitope in Saade, G., Deblanc, C., Bougon, J., Marois-Créhan, C., Fablet, C., Auray, G., et al. (2020). its N-terminal ectodomain. J. Gen. Virol. 84, 1535–1543. doi: 10.1099/vir. Coinfections and their molecular consequences in the porcine respiratory tract. Vet. Res. 0.18957-0 51, 80–19. doi: 10.1186/S13567-020-00807-8 Worobey, M., Pekar, J., Larsen, B. B., Nelson, M. I., Hill, V., Joy, J. B., et al. (2020). The Shi, M., Holmes, E. C., Brar, M. S., and Leung, F. C.-C. (2013). Recombination is emergence of SARS-CoV-2 in Europe and North America. Science 370, 564–570. doi: associated with an outbreak of novel highly pathogenic porcine reproductive and 10.1126/SCIENCE.ABC8169 respiratory syndrome viruses in China. J. Virol. 87, 10904–10907. doi: 10.1128/ Yoon, I. J., Joo, H. S., Goyal, S. M., and Molitor, T. W. (1994). A modified serum JVI.01270-13 neutralization test for the detection of antibody to porcine reproductive and respiratory Shi, M., Tsan-Yuk Lam, T., Hon, C.-C., Kin-Hei Hui, R., Faaberg, K. S., Wennblom, T., syndrome virus in swine sera. J. Vet. Diagn. Investig. 6, 289–292. doi: et al. (2010). Molecular epidemiology of PRRSV: a phylogenetic perspective. Virus Res. 154, 10.1177/104063879400600326 7–17. doi: 10.1016/j.virusres.2010.08.014 Zhao, D., Yang, B., Yuan, X., Shen, C., Zhang, D., Shi, X., et al. (2021). Advanced research Snijder, E. J., and Meulenberg, J. J. (1998). The molecular biology of arteriviruses. J. Gen. in porcine reproductive and respiratory syndrome virus co-infection with other pathogens Virol. 79, 961–979. doi: 10.1099/0022-1317-79-5-961 in swine. Front. Vet. Sci. 8:699561. doi: 10.3389/FVETS.2021.699561 Frontiers in Microbiology 14 frontiersin.org