Molecular Characterization of Porcine Reproductive and Respiratory Syndrome Virus in Korea from 2018 to 2022

Porcine reproductive and respiratory syndrome (PRRS) is an endemic disease in the Republic of Korea. Surveillance of PRRS virus (PRRSV) types is critical to tailor control measures. This study collected 5062 serum and tissue samples between 2018 and 2022. Open reading frame 5 (ORF5) sequences suggest that subgroup A (42%) was predominant, followed by lineage 1 (21%), lineage 5 (14%), lineage Korea C (LKC) (9%), lineage Korea B (LKB) (6%), and subtype 1C (5%). Highly virulent lineages 1 (NADC30/34/MN184) and 8 were also detected. These viruses typically mutate or recombine with other viruses. ORF5 and non-structural protein 2 (NSP2) deletion patterns were less variable in the PRRSV-1. Several strains belonging to PRRSV-2 showed differences in NSP2 deletion and ORF5 sequences. Similar vaccine-like isolates to the PRRSV-1 subtype 1C and PRRSV-2 lineage 5 were also found. The virus is evolving independently in the field and has eluded vaccine protection. The current vaccine that is used in Korea offers only modest or limited heterologous protection. Ongoing surveillance to identify the current virus strain in circulation is necessary to design a vaccine. A systemic immunization program with region-specific vaccinations and stringent biosecurity measures is required to reduce PRRSV infections in the Republic of Korea.


Introduction
Porcine reproductive respiratory syndrome (PRRS) is an infectious disease that contributes to massive economic losses in the swine industry. The PRRS virus (PRRSV) is responsible for reproductive failure in mature pigs and respiratory disease in pigs of all ages [1][2][3]. The virus belongs to the family Arteriviridae and the order Nidovirales and has two genotypes, PRRSV-1 (European type) and PRRSV-2 (North American type), which have been reclassified as Betaarterivirus suid 1 and Betaarterivirus suid 2, respectively [4]. They share 60% genomic identity and 20% nucleotide sequence variability within each genotype [5,6].
Among the NSPs, non-structural protein 2 (NSP2) is a diverse viral protein that is important in the life cycle of PRRSV. It primarily expresses proteases involved in host immune response and viral replication. It has the highest genetic diversity and is used as a molecular marker to study the molecular epidemiology and evolution of PRRSV [12,13].

Phylognetic Analysis Based on ORF5
Phylogenetic analysis was performed by using the CLC Workbench tool (QIAGEN, Aarhus A/S. Aarhus, Denmark). The phylogenetic trees were constructed via the neighborjoining method with the Jukes-Cantor method for nucleotide distance and a bootstrap value of 1000 on ORF5 nucleotide sequence data (738 PRRSV-2 and 37 PRRSV-2 reference, 668 PRRSV-1 and 14 PRRSV-1 reference). Nucleotide and amino acid sequence identities were calculated via pairwise comparison of sequences on the CLC Workbench.

Analysis of Amino Acids of PRRSV
The amino acid sequences of the samples were aligned by using a CLC Workbench. The sequence logos of the samples were compared with those of the reference strains. Amino acid variations in the neutralizing epitope, B-cell epitopes, T-cell epitope, and hypervariable region were analyzed.

Analysis NSP2 Gene
RT-PCR targeting NSP2 was performed by using forward and reverse primer sets. The amplified NSP2 gene was sequenced for further analysis. A total of 258 sequences (130 PRRSV-1 and 128 PRRSV-2) was aligned with the reference strains (three PRRSV-1 and nine PRRSV-2 references) on the CLC Workbench. Aligned sequences were extracted and translated into amino acid sequences. Aligned amino acid sequences were analyzed for deletion or insertion patterns via comparison with the reference sequences.

Phylogenetic Analysis
The PRRSV-1-positive samples were divided into three subgroups (A, B, and C) under subtype 1. Of the 720 sequences in the PRRSV-1 group, 89% (639) were subtype 1A,

Phylogenetic Analysis
The PRRSV-1-positive samples were divided into three subgroups (A, B, and C) under subtype 1. Of the 720 sequences in the PRRSV-1 group, 89% (639) were subtype 1A,

Analysis of Amino Acids of PRRSV-2 GP5 (ORF5 Gene) Protein
The

Non-Structural Protein (NSP2) Deletion Pattern of PRRSV-1
NSP2 sequences of PRRSV-1 showed a typical 19 amino acid deletion pattern at aa positions 361-379, similar to the E38 (subgroup A reference strain). In addition to the standard 19 amino acid deletion, an additional 11 amino acids (in 16 sequences) from positions at aa 418-428 and 3 amino acids (in 15 sequences) at aa positions 355-357 were observed (Figure 7).

Non-Structural Protein (NSP2) Deletion Pattern of PRRSV-2
NSP2 sequences of PRRSV-2 were aligned and analyzed on the CLC Workbench. A total of 39 sequences did not show amino acid deletion, similar to the reference strain VR2332. In total, 4 sequences were similar to lineage 8 but had a 111 aa discontinuous deletion pattern, and 1 more sequence matched lineage 8 and showed a 6 aa deletion at aa positions 201-206. The rest of the NSP2 sequences showed a typical 131 aa discontinuous deletion pattern (111aa + 1aa + 19aa) similar to the reference strains NADC30-like, LKA, LKB, or LKC. In total, 2 of the sequences that showed 131 amino acid deletion also showed an additional amino acid deletion at aa positions 162-164, and 3 more sequences showed 5 additional amino acid deletions at aa positions 158-162 (Figure 8).
The ORF5 sequences of the 1411 PRRSV isolates were analyzed and classified into several subtypes and lineages based on their phylogenetic relationships. Most PRRSV-1 samples evaluated between 2018 and 2022 belonged to subgroup A (89%). In fact, the majority of the samples analyzed, including those for PRRSV-1 and PRRSV-2, belonged to subgroup A (42%). The subgroup C population was minor, and only one subgroup B isolate tested positive for ORF5. Similar observations were made by Kim et al. [27], who found that subgroup A consistently remained in the majority of PRRSV outbreaks from 2014 to 2019.
All PRRSV-1 isolates analyzed for the NSP2 deletion pattern showed a typical 19 amino acid deletion at aa positions 361 to 379, which is similar to that of the E38 (KT033457.1) reference strain, which belongs to subgroup A. Apart from the typical 19 amino acid deletion pattern, a few isolates showed 3 amino acid deletion at aa positions 355-357, and a few others showed an 11 amino acid deletion from aa positions 418 to 428. Two amino acids, 418 and 420, are part of the ES4 epitope [47] which may lead to immune evasion and the development of distinct viral isolates. These isolates require further investigation to determine the effect of this deletion.
Subgroup A has a nucleotide similarity of 79.87-88.78% with VP-046 (UNISTRAIN PRRS, HIPRA, GIRONA, Spain), 82.54-88.12% with DV (Porcilis PRRS, MSD, the Netherlands). Subgroup C shares 86.14-100% and 90.26-99.83% similarity with the DV and VP-046 vaccine strains sequence, respectively. According to a nucleic acid analysis study, the vaccine does not completely protect animals against subgroup A infections. The PRRSV-1 vaccine used in Korean swine farms, which contains DV and or VP-046 strains, is unable to protect animals from subgroup 1A infection, and these modified live viruses revert to virulent forms, causing infection [27,48]. Antibodies exert strong positive selection pressure on PRRSV by targeting specific viral subpopulations while allowing the establishment of other subpopulations. Vaccination against the PRRSV results in genetic heterogeneity [49]. Subgroup B isolates were grouped with Thai (03RB1) and Danish strains (361-364), whereas subgroup C isolates were grouped with South Korean vaccine strains (DV MSD and VP-046 HIPRA) [50]. Immune evasion and diversifying evolution of Korean PRRSV-1 subgroup A field isolates against sequencing suggest that subgroup A is evolving independently and establishing in the Republic of Korea.
Current vaccines used in Korea contain virus strains from subgroup C, which are insufficient to protect animals affected by subgroup A infections. Therefore, a subgroup-Aspecific vaccine that includes local isolates is required.
PRRSV-2 causes more severe respiratory illnesses than PRRSV-1 isolates [51]. Lineage 1 accounted for 40% (322/802) of the PRRSV-2 from 2018 to 2022. The prevalence of lineage 1 showed an increasing trend from 2018 to 2022, whereas lineage 5, LKB, and LKC showed a decreasing trend. Kim et al. [27] observed a similar pattern in their temporal dynamic study of PRRSV in Korea from 2014 to 2019. Lineage 1 is widespread in the United States and Canada. Lineage 1 may have spread owing to swine trading and artificial insemination in Korea, China, and Taiwan [27,38,46,52]. The dominance of Lineage 1 in Korea is concerning because it contains some of the most pathogenic strains, including NADC30-like, NADC34like, and MN184 strains.
Lineage 1 was the second largest population (29.6%) of PRRSV in 2019 in Korea [27], and in Peru, 75% of the strains detected from 2015 to 2017 were NADC34-like strains [53]. According to Xu et al. [54], lineage 1 (1.5 and 1.8) accounted for 64% of positive samples in 2021. Lineage 1 strains are the most common in the Canadian provinces of Ontario and Quebec [53]. Global vaccine containing lineage 5 provided partial protection against lineage 1. It is ideal to regularly monitor PRRSV and design safe and effective vaccines that are based on current circulating strains.
This study identified 40 highly pathogenic strains that were similar to NADC30-like, NADC34-like, and MN184 strains. Since 2005, the MN184 strain has emerged and spread throughout the Korean pig population [55]. NADC30-like strains emerged in Korea in 2015, whereas non-NADC30-like strains emerged in 2017 [23]. Since 2005, MN184-like strains have emerged and become widespread in Korean pigs [55]. The growing number of highly pathogenic strains is a major concern because they are frequently involved in recombination with other PRRSV strains. Kim et al. [23] observed lineage 1 recombinants with NADC34 as the major parent and NADC30 as the minor parent, with recombination signals in the NSP2 and NSP10 regions. According to our findings, many isolates were similar to NADC34-like strains based on the ORF5 sequence, but only three sequences were similar to those of NADC30-like virus isolates. Based on the NSP2 amino acid deletion pattern, none of the viruses resembled NADC34 in this study. However, the genomes of these isolates require further investigation. Lineage 1 isolates required recombination analysis to identify potential recombinants.
In the current study, LKB and LKC were the next most prevalent PRRSV (2018-2022). LKA, LKB, and LKC were identified in 2003, 2014 [14], and 2005, respectively [56,57]. The LKA, LKB, and LKC lineages have developed genetic components that are geographically distinct from those of the Republic of Korea, the origins of which are still unknown. ORF5 phylogenetic analysis distinguished LKA, LKB, and LKC into three distinct clusters. In contrast, whole-genome sequencing and NSP2 phylogeny merged into one large branch with two sub-branches, indicating that they share a common ancestor.
Most Korean farms use Ingelvac PRRS modified-live vaccine (MLV) (PRRSV-2 lineage 5), which was introduced in 1995. We identified vaccine-like strains, and while the percentage of lineage 5 decreased, these vaccine-like isolates remain a concern in the field. Kim et al. [23] reported a vaccine-like strain and LKC recombination, and it was suspected that LKB was generated by the recombination of LKC and MLV strains (Ingelvac). This is supported by findings from other studies that indicate that the use of modified live viruses contributes to increased PRRSV genetic diversity [58][59][60][61]. To avoid the emergence of new viral strains or lineages, modified live viruses must be used with caution.
Although there are fewer lineage 8 isolates, they are very important because they are classified as highly pathogenic strains [62]. In 2014, a modified live virus vaccine from Fostera PRRSV, which is an attenuated isolate of lineage 8 (P129 strain), was introduced in Korea. The origin of lineage 8 in Korean swine farms could be attributed to the introduction of the Zoetis Fostera vaccine or the importation of pigs from other countries with lineage 8 prevalence. There were seven lineage-8-like isolates based on ORF5, and four of these isolates had 111 discontinuous amino acid deletion patterns of NSP2, which is very similar to that of lineage 1 (lineage 1 shows a continuous 111 deletion pattern). Such samples require further investigation by using whole-genome sequencing.
According to Kwon et al. [31], the NSP2 amino acid deletion pattern is similar to that of the MN184-like strain, but it has a high level of nucleotide identity with VR-2332 in the ORF5-ORF7 sequence. Evidence for the possible involvement of recombination in field PRRSV evolution was first documented in the United States in 1996 and later in China [38,[63][64][65]. Recombination events occur between vaccine and field strains [23,66]. Although no recombination analyses were performed in this study, it would be beneficial to examine these samples for recombination in the future.

Conclusions
From 2018 to 2022, the majority of the samples in the Republic of Korea belonged to subgroup A (42%), followed by lineages 1 (21%) and 5 (14%) and LKC (9%), LKB (6%), and subgroup C (5%). There were only a few samples from lineage 8, LKA, and subgroup B. The increase in the prevalence of subgroup A (PRRSV-1) and lineage 1 was a cause for concern. Currently available MLV vaccines contain subgroup C (PRRSV-1) and lineage 5 VR-2332 (PRRSV-2) viruses, but they do not completely protect against the more common subgroup A or lineage 1 viruses. It is well known that in case of PRRSV, there will be far less heterologous protection, indicating that the vaccine cannot protect the animals from another existing lineage. Instead, the number of infections due to subgroup A, lineage 1, LKB, and LKC increased, indicating that these isolates escaped the immunity established by vaccination with lineage 5 or subgroup C. Therefore, it is critical to develop an improved and safe vaccine that includes the prevalent type/lineage and does not mutate or recombine with field-circulating viruses. Therefore, continuous monitoring and strengthening of PRRSV prevention and control are necessary. In addition to vaccination, biosecurity measures such as restricted entry of people, supplies, and vehicles; air filtration; manure management; disposal of dead bodies; establishing PRRSV-negative boar studs and gilt sources; and careful introduction of gilts into the herd must be implemented.